OBJECTIVITY IN THE NATURAL SCIENCES
Our penchant for simplification and the compelling notion that the world is governed by finite set of fundamental rules leads us to reductionism. A reductive approach allows focused investigation of the components of phenomena otherwise daunting in their complexity. Reduction in the sciences has been a spectacular success. We can say that biology rests on a foundation of chemistry which in turn lies on the bedrock of physics.
EMERGENT PROPERTIES
As we transition from one level in the traditional knowledge hierarchy to the next, new phenomena always emerge. Emergent properties are, by definition, not predictable in advance from a reductionist vantage point at a previous, more fundamental level.
Coupling reductionism with a holistic perspective recognizes that the synergistic whole is greater than the sum of the parts.
In constructing their theories, scientists seek explanatory and predictive power in the real world. Predictability in science rests on inductive reasoning, whereby a generalization is made from a set of particular instances. Induction is inextricable from how we encounter a more or less uniform world. Although we must concede that inductive reasoning is psychological rather than logical, the criterion of falsifiability eliminates the nagging doubt that the entire scientific edifice lacks a rigorous, logical foundation.
SCIENCE IS EXPERIMENTAL
Scientists test falsifiable hypotheses with controlled experiments. Experimental protocols are refined to minimize sources of error and to ensure the repeatability and the statistical validity of results. Any distracting anomalies are isolated. Any revelatory anomalies are seized upon and examined carefully. They may steer work in new directions.
Data analysis involves searching for points of interest, patterns and connections to expose correlation. But correlation is not causation. If convincing evidence for underlying causation is established, scientists know that emergent factoids and scientific explanations, whether large or small, are never final. They are refutable, or falsifiable, by design and, therefore, tentative.
SCIENTIFIC MODELS
Models are metaphorical ways of describing, predicting and understanding the natural world. They are the all-too-human, pragmatic response to the fact that we can never encapsulate the complexity and connectivity of nature in its entirety.
Scientific models are fictive simplifications with predictive power. When we run a scientific model—whether cognitively (in our heads) or mathematically (usually on a computer)—real world details that fall outside the defined parameters of the model are assumed to be irrelevant.
Models are ways of thinking. We can represent them with diagrams, formulae and three-dimensional sculptures. The representations are not the models themselves.
We might imagine a model with so many complex features that it would truly reflect and predict the entire universe; but this would be a virtual replica rather than a model since it would entail one-to-one mapping with nature itself.
