The Astrophysics Spectator

Home

Topics

Interactive Pages

Commentary

Other Pages

Information

Commentary

The Nature of Theory

A bad observer is wrong 5% of the time, a good theorist is right 5% of the time. This joke has floated around the astrophysical community for years. It reflects the complexity of astronomical phenomena and the large number of choices that face the theorist when he constructs a theory, but it also often reflects a fundamental misunderstanding about the nature of theoretical research. The joke implies that the theorist is a soothsayer, a person who can divine the universe through pure reason, but with less than stellar accuracy.

Perhaps this view of theory is inevitable. Logic and mathematics preceded formal science by centuries, suggesting that these two disciplines come to us more naturally than does the discipline of testing theories and discarding those that do not match experimental results. The history of astronomy has numerous examples of scientists vainly clinging to to theories that are at odds with the observations.

We have the example of four decades ago of two theories of cosmology, the cold cosmology and the steady state cosmology, that persisted far longer than the observations warranted. The cold cosmology is an expanding universe cosmology based on the idea that the early universe show have a low entropy. The early universe is in a cold solid state, in contrast to the high temperature state of the early universe in the big bang cosmology. The steady state cosmology is based on the idea that the appearance of the universe should not change with time, so as the universe expands, mater spontaneously comes into existence to keep the density of the universe constant. These two cosmologies do not predict a microwave background, in contrast to the big bang cosmology, which does. The discovery of the microwave background should have immediately buried these theories, but adherents modified these theories in ad hoc to account for the initial observations. The inability of these ad hoc explanations to reproduce the results of later, more refined experiments eventually undermined all support for these theories.

We have the example of a prediction last decade that the decay of a hypothetical fundamental particle would be observed in the space space at ultraviolet wavelengths. When experiments failed to find the expected signature, the theorist who predicted the decay of this particle immediately offered an explanation of why the decay is not observable, while maintaining that the hypothetical particle may exist.

This phenomena is not recent; it goes back to the origins of astronomy when our forefathers insisted that all planetary orbits were circle. As they gradually found disagreement between the actual planetary motion and the motion they expected, they added circular epicycles to their circular orbits, and then circular epicycles to their epicycles. Effectively they were reproducing the true motion through an infinite series of circular motions, all motivated by their inability to discard their notion that only circular motion is possible.

These instances are common enough that one saying within the astronomical community is that science advances when the current generation dies. It is human nature to preserve our world view in the face of evidence to the contrary, particularly if that world view comes after years of effort, and if our prestige rests upon it. Objectivity is a great act of will, and science's demand for a ruthless objectivity is a difficult standard for any scientist to meet. No one wants to be wrong, even if being wrong 95% of the time is considered doing well.

This tendency to cling to a world view is tied to how we understand the structure of the world. Our mind naturally reduces the world around us to a set of basic principles, and then categorizes objects and events in terms of that world view. This is the reason that we stop and stare when we see something unexpected: the mind must spend more time analysing the visual information to understand what it is seeing.

When we construct a new theory, we are inverting the process of constructing a world view. We are selecting a small set of principles and constructing a theory based on those principles. We have an underlying set of principles for the theory, and we intuitively feel that the theory must therefore be correct. Our predisposition to use a world view gives us the notion that we can understand the nature of the universe through pure reason.

The philosophy underlying science is hostile to this viewpoint. Science is the objective acquiring of knowledge, and for a theory to be correct, it must meet the objective standards of science. The act of constructing a theory from a priori principles is a purely subjective act. The aesthetic sense that guided our choices in constructing a theory is derived from the evolutionary requirement that we survive as a species. Why should this aesthetic sense lead us a priori to the correct theory for an x-ray binary system, a fundamental particle, or the expansion of the universe?

What is the nature of theory? Alone it is subjective, and a priori it cannot cannot presage experimental results. Its role in science is to present alternatives that are distinguishable through experiment. Objectively, a health theoretical community hopes that most of its theories are proven invalid, because it is through the elimination of plausible alternative theories that scientific understanding advances. The good theorist is not wrong 95% of the time, rather his theories are experimentally eliminate 95% of the time, and in a universe as complex as ours, this is a considerable achievement.

Jim Brainerd

Ad image for The Astrophysics Spectator.