The Philosophy of Science

The solar eclipse of May 29, 1919, forced a rethink of fundamental laws of physics

By Keith Tidman

Science aims at uncovering what is true. And it is equipped with all the tools — natural laws, methods, technologies, mathematics — that it needs to succeed. Indeed, in many ways, science works exquisitely. But does science ever actually arrive at reality? Or is science, despite its persuasiveness, paradoxically consigned to forever wending closer to its goal, yet not quite arriving — as theories are either amended to fit new findings, or they have to be replaced outright?

It is the case that science relies on observation — especially measurement. Observation confirms and grounds the validity of contending models of reality, empowering critical analysis to probe the details. The role of analysis is to scrutinise a theory’s scaffolding, to better visualise the coherent whole, broadening and deepening what is understood of the natural world. To these aims, science, at its best, has a knack for abiding by the ‘laws of parsimony’ of Occam’s razor — describing complexity as simply as possible, with the fewest suppositions to get the job done.

To be clear, other fields attempt this self-scrutiny and rigour, too, in one manner or another, as they fuel humanity’s flame of creative discovery and invention. They include history, languages, aesthetics, rhetoric, ethics, anthropology, law, religion, and of course philosophy, among others. But just as these fields are unique in their mission (oriented in the present) and their vision (oriented in the future), so is science — the latter heralding a physical world thought to be rational.

Accordingly, in science, theories should agree with evidence-informed, objective observations. Results should be replicated every time that tests and observations are run, confirming predictions. This bottom-up process is driven by what is called inductive reasoning: where a general principle — a conclusion, like an explanatory theory — is derived from multiple observations in which a pattern is discerned. An example of inductive reasoning at its best is Newton’s Third Law of Motion, which states that for every action (force) there is an equal and opposite reaction. It is a law that has worked unfailingly in uncountable instances.

But such successes do not eliminate inductive reasoning’s sliver of vulnerability. Karl Popper, the 20th-century Austrian-British philosopher of science, considered all scientific knowledge to be provisional. He illustrated his point with the example of a person who, having seen only white swans, concludes all swans are white. However, the person later discovers a black swan, an event conclusively rebutting the universality of white swans. Of course, abandoning this latter principle has little consequence. But what if an exception to Newton’s universal law governing action and reaction were to appear, instead?

Perhaps, as Popper suggests, truth, scientific and otherwise, should therefore only ever be parsed as partial or incomplete, where hypotheses offer different truth-values. Our striving for unconditional truth being a task in the making. This is of particular relevance in complex areas: like the nature of being and existence (ontology); or of universal concepts, transcendental ideas, metaphysics, and the fundamentals of what we think we know and understand (epistemology). (Areas also known to attempt to reveal the truth of unobserved things.) 

And so, Popper introduced a new test of truth: ‘falsifiability’. That is, all scientific assertions should be subjected to the test of being proven false — the opposite of seeking confirmation. Einstein, too, was more interested in whether experiments disagreed with his bold conjectures, as such experiments would render his theories invalid — rather than merely provide further evidence for them.

Nonetheless, as human nature would have it, Einstein was jubilant when his prediction that massive objects bend light was confirmed by astronomical observations of light passing close to the sun during the total solar eclipse of 1919, the observation thereby requiring revision of Newton’s formulation of the laws of gravity.

Testability is also central to another aspect of epistemology. That is, to draw a line between true science — whose predictions are subject to rigorous falsification and thus potential disproof — and pseudoscience — seen as speculative, untestable predictions relying on uncontested dogma. Pseudoscience balances precariously, depending as it does on adopters’ fickle belief-commitment rather than on rigorous tests and critical analyses.

On the plus side, if theories are not successfully falsified despite earnest efforts to do so, the claims may have a greater chance of turning out true. Well, at least until new information surfaces to force change to a model. Or, until ingenious thought experiments and insights lead to the sweeping replacement of a theory. Or, until investigation explains how to merge models formerly considered defyingly unalike, yet valid in their respective domains. An example of this last point is the case of general relativity and quantum mechanics, which have remained irreconcilable in describing reality (in matters ranging from spacetime to gravity), despite physicists’ attempts. 

As to the wholesale switching out of scientific theories, it may appear compelling to make the switch, based on accumulated new findings or the sense that the old theory has major fault lines, suggesting it has run its useful course. The 20th-century American philosopher of science, Thomas Kuhn, was influential in this regard, coining the formative expression ‘paradigm shift’. The shift occurs when a new scientific theory replaces its problem-ridden predecessor, based on a consensus among scientists that the new theory (paradigm) better describes the world, offering a ‘revolutionarily’ different understanding that requires a shift in fundamental concepts.

Among the great paradigm shifts of history are Copernicus’s sun-centered (heliocentric) model of planet rotation, replacing Ptolemy’s Earth-centered model. Another was Charles Darwin’s theory of natural selection as key to the biological sciences, informing the origins and evolution of species. Additionally, Einstein’s theories of relativity ushered in major changes to Newton’s understanding of the physical universe. Also significant was recognition that plate tectonics explain large-scale geologic change. Significant, too, was development by Neils Bohr and others of quantum mechanics, replacing classical mechanics at microscopic scales. The story of paradigm shifts is long and continues.

Science’s progress in unveiling the universe’s mysteries entails dynamic processes: One is the enduring sustainability of theories, seemingly etched in stone, that hold up under unsparing tests of verification and falsification. Another is implementation of amendments as contrary findings chip away at the efficacy of models. But then another is the revolutionarily replacement of scientific models as legacy theories become frail and fail. Reasons for belief in the methods of positivism. 

In 1960, the physicist Eugene Wigner wrote what became a famous paper in philosophy and other circles, coining the evocative expression unreasonable effectiveness. This was in reference to the role of mathematics in the natural sciences, but he could well have been speaking of the role of science itself in acquiring understanding of the world.