The onion of Science (2002)

I often have people ask me what I do for a living. Because my research projects are so varied I like to give a different answer each time. Nevertheless, the question has forced me to evaluate my role as a scientist. This page is an attempt to explain what I believe science is all about and where my work fits in…

Background

Science is the process of making testable claims (scientists call them theories or hypotheses) about our surroundings and testing them. When a claim fails a test we are forced to re-evaluate either it or the assumptions it relies on. On the other hand, the more tests a claim passes the more it is accepted by the scientific community. (Or at least it should be, but sometimes other factors can affect how quickly a claim is accepted, such as culture, religion or politics.)

By natural extension, then, the ultimate scientific achievement would be to propose a claim (theory) which can explain everything! Given that, one might reasonably ask “Why aren’t all scientists looking for a Theory of Everything?”

Onion

In fact, some physicists are trying to find the Theory of Everything. But besides that we have many other scientific fields, such as chemistry, biology, psychology, etc. Why is that? Each field covers only a small domain of knowledge. It tends to begin with some axiomatic principles and explore those to form theories to explain more complicated phenomena. For example some physicists study how sub-atomic particles interact to form atoms. Further simple examples follow:

The important point is that each scientific field has a limited domain of expertise; it can explain only a portion of reality. It is not (currently) possible, for example, to explain why people smile from a knowledge of quarks and gluons alone.

Notice the relationship between the axioms and theories in the above table. The concepts developed in one field often become the fundamental principles upon which another field rests. I have found it convenient to think of the relationships between scientific fields as separate layers of knowledge surrounding a central core. One layer forms the foundation for another. Hence, the onion model of science.

Each layer of the onion represents a field of science which is tightly connected, and the lines indicate the limits of the field. Of course, this is an idealization: there are many fields which overlap each other, blurring the separation between layers. (For instance, cell membranes are studied by physicists, chemists and biologists.) However, the onion is meant to emphasize the strong interconnectedness within fields and neglect the (relatively) few links between fields.

The reader may ask why is it possible to separate science into layers? Why don’t the layers just blend together into a cohesive whole? The answer has to do with how science is typically done.

Reductionism

If you try to take a cat apart to see how it works, the first thing you have on your hands is a nonworking cat.

— Douglas Adams, 2002

The separation of science into layers is not due to incomplete knowledge in the core but rather due to the historical methodology of science.

Typically, a system is studied by taking it apart and seeing how its individual components work, a method called reductionism. It’s a very natural and intuitive approach to learning and it usually works. But sometimes, even when you know all there is to know about all the individual bits, you still can’t figure out how the system functions as a whole. Consider a mechanical watch: it consists primarily of simple components, such as cogs and springs, and yet few people would be able to explain how it works.

The difficulty is that, to understand it, you have to think about how all the bits are working together at the same time. The bits are said to be working in parallel. Nature, of course, is massively parallel so we can’t expect a reductionist approach to work every time.

I am not trying to belittle reductionism, it is a very useful technique and has had many astounding successes. Without it no progress in science could have been possible whatsoever. My point is that, as good as reductionism is, it cannot explain everything. Even if a Theory of Everything is discovered which perfectly explains every question that could be asked at the core of the onion, it does not follow that we will know everything that science can ask. There will still be a limit to its explanatory power. A complete understanding of physics will not impact biology, for example. But if the separation of science into distinct layers (or fields) is not due to incomplete understanding then what is it due to?

Complexity

As alluded to with the analogy of the watch, the problem is the sheer numbers of interacting components in one layer make it impossible to compute results in a higher layer of the onion. Thus, a Theory of Everything will not spell the end of science, it will just fill in the rules at the very core of the onion. By itself it will contribute nothing to the understanding of the higher layers.

To understand nature as a whole it is necessary to construct new theories and tools to explain large systems of interacting units. This is the science of Complexity. Complexity is a relatively new approach to science which complements reductionism. Reductionism tries to explain how each of the bits work and complexity tries to explain how they all work with each other. Together, these two approaches hold the promise of full and complete understanding of the natural world.

As complexity science matures it is hoped that it will give us a way to break the barriers separating the layers of the onion (fields of science). As such complexity science is interdisciplinary, it is not tied to any particular field of study. In fact, it can be found in fields as diverse as condensed matter physics and population ecology.

— Rik Blok, 2002