In fact, science is a method by which knowledge about the physical world is attained. Scientific 'knowledge' inheres both observed, empirical data and the best possible explanation regarding the mechanisms by which the observed phenomena came about and the prediction, where appropriate, of any future manifestations that can reasonably be expected. The scientific method is most appropriately applied to natural, physical phenomena, so it defines the subject areas appropriate to investigation.
Scientific method, methodological naturalism, is akin to formalized skepticism in that it ideally proceeds by rigorous scrutiny of falsifiable hypotheses. To achieve this, postulated explanations for observed phenomena must be expressed in such a way that they can be tested and disproven. An analogy would be determining whether a suspect in a crime has an alibi – if the suspect can be demonstrated to have been in Beijing at the time that a stabbing was committed in San Francisco, then the suspect cannot have perpetrated the stabbing (unless, that is, the suspect had impossibly long arms).
Any hypotheses that are demonstrated to be faulty will be discarded, and alternate explanations for empirical observations will be formulated and tested. (Expanding the analogy, another suspect will be sought for the stabbing.) Eventually, any reasonable hypotheses that are not disproven will be regarded as so acceptable as to be elevated to the level of scientific theory.
Scientific predictions, however, represent a subset of experimentation and are propositional – if this hypothesis is correct, then we will observe such and such a phenomenon. Failure to observe the predicted phenomenon might be taken to disprove the hypothesis. However, the failure might be a result of experimental or observational error, or might result from faulty predictions based upon a reasonable hypothesis. Alternatively, the hypothesis may be incorrect, but the predicted phenomenon is observed because of a mechanism not yet hypothetically considered.
For these reasons peer-reviewed scientific papers include analyses of current thinking, descriptions of methods, and statements of results so that other researchers might evaluate conclusions and attempt replication or alternate experimentation. In science, unlike the case for mathematics, proof is not possible, while disproof – falsification – is possible. For this reason, hypotheses to be experimentally tested are ideally framed in such a way that they may be disproved – falsifiable hypotheses. When an empirically-based, logical hypothesis, which has not been disproved after repeated testing, is deemed satisfactory by consensus within the scientific community, then the hypothesis graduates to the status of Theory (capitalized to differentiate the scientific term from its vernacular usage).
In practice, much of science proceeds upon positive results – repeated observations of a phenomenon under particular conditions. In the softer sciences, such as the social sciences, statistical analyses of results play an important role. Some sciences, such as paleontology are by their nature outside the possibility of experimentation – we cannot resurrect dinosaurs or recreate meteor impacts – and must proceed on the basis of accumulated empirical evidence.
We all toss around vernacular 'theories' – unproved ideas or theoretical speculations that are not necessarily even so well formulated as scientific hypotheses. Even some elaborate theories, such as intelligent [sick] design theory have considerably less conceptual merit than their promoters will ever acknowledge. For biblical literalists, such a 'theory' may have emotional appeal, but in terms of speculation concerning reality, ‘intelligent [sick] design theory’ has no more merit than the belief system of the Lambayeques.
Ȣ biological evolution Ȣ evolution Ȣ mutation Ȣ natural selection
å Fundamentals å On Further Thought