We often hear claims about what is or isn’t “scientific.” Science holds a distinctive place in modern society; we trust its pronouncements on everything from building bridges to who counts as an expert. Yet, history is full of scientific theories that turned out to be wrong. This leads us to a fundamental question: what exactly is it about science that makes it special? And perhaps even more importantly, how do we distinguish genuine science from something that might claim to be science but isn’t – what philosophers call pseudoscience?
This question, known as the problem of demarcation, is of enormous practical importance, influencing decisions about everything from education to public policy.
Much of the modern discussion about the philosophy of science, including the problem of demarcation, has its roots in the logical positivism (or logical empiricism) movement that emerged in early 20th-century Vienna. Inspired by the revolutionary work of Albert Einstein, the positivists were interested in setting philosophy on a more empirical footing, focusing on logic, language, and what could be tied back to sensory experience. For them, science was special because it dealt with “fact-stating discourse,” while attempts to state “extra-scientific facts” amounted to metaphysics, which they viewed as meaningless. They sought criteria to distinguish scientifically meaningful statements from meaningless metaphysical ones, a key concept being the verification principle, which suggested a statement was meaningful if verifiable by possible observations.
However, observation, as Popper noted, can be “cheap,” essentially interpretation in the light of theory. This points to a challenge in relying solely on confirmation by evidence as the mark of science.
Popper’s Provocative Idea: Falsification
A contemporary of the positivists in early 20th-century Vienna, Karl Popper, tackled the demarcation problem directly. Popper was particularly struck by the difference between Einstein’s theory of relativity and theories like those of Sigmund Freud, Alfred Adler, and Karl Marx. While many believed the latter theories were genuinely scientific, Popper became disenchanted with them. He observed that these theories seemed able to explain any potential observation, turning apparent counterevidence into support. Freudians, for example, had “ready explanations for any observational result.” Pseudoscientists, in Popper’s view, find “confirming evidence everywhere.” This made their claims essentially immune to being proven wrong.
Popper’s radical proposal was that the distinguishing mark of science is not that it is confirmed by evidence, but that it is open to criticism and seeks to falsify its hypotheses. Genuine scientific theories, like Einstein’s, make risky predictions that could, in principle, show the theory to be false. For Popper, the most we can say in favor of a theory is that it has survived strenuous attempts to falsify it, a state he called corroboration. Importantly, corroboration does not mean the theory is likely true, only that it hasn’t been proven false yet.
Thus, for Popper, theories were pseudoscientific if they were formulated in a way that made them immune to refutation.
Challenges to Falsification and Other Criteria
While elegant, Popper’s falsification criterion faces significant challenges. It’s surprisingly difficult to interpret in a way that clearly separates all of physics from, say, Freudian theory. Real sciences, like medicine, tolerate quite a number of false predictions without being rejected. Complex scientific theories rely on many auxiliary hypotheses, meaning an empirical test that fails only shows that something is wrong somewhere in the web of beliefs, not necessarily the core theory being tested. We cannot require a theory to be rejected simply due to “persistent failures of fit with the evidence,” as this would leave us with very little science. It’s hard to precisely characterize the difference in falsifiability between fields like astrology and physics. Moreover, it’s not clear that scientists actually do or should reject theories immediately upon encountering conflicting results.
Philosophers have proposed other criteria for demarcation, but each also faces problems.
- Progress—One historical criterion is that pseudosciences tend not to make much progress compared to rival theories. However, progress itself is tricky to characterize and measure. A theory’s scientific status might even change over time based on its competitors’ progress.
- Lack of mechanism—Some pseudosciences, like astrology, lack a clear mechanism for their claimed effects. But historically, even legitimate sciences, like Newton’s theory of gravity, were accepted without a known physical mechanism for action at a distance.
- Social practice—Defining science based on institutional acceptance (journals, universities) seems problematic, as it would count politically institutionalized but scientifically flawed systems, like Lysenkoist biology, as scientific.
- Dubious origins—Many genuine sciences, including chemistry (from alchemy), arose from epistemically questionable origins. Their origins don’t preclude them from later achieving scientific status.
- Forms of reasoning—It’s difficult to show that specific forms of reasoning definitively distinguish science from pseudoscience, as both may use mathematical or causal inferences, and science sometimes employs reasoning strategies also found in pseudosciences.
The Case of Creationism
The debate over creationism (including Young-Earth Creationism and Intelligent Design Creationism) highlights the difficulty of finding a clean demarcation criterion. Critics argue that creationism often relies on negative arguments against established science (like evolution) rather than developing its own positive, testable proposals. While one can have a valuable scientific career trying to falsify a theory, the status of an entire discipline might depend on its positive contributions. Intelligent Design, for example, has been noted for being dominated by big, overarching questions rather than the “little questions that can be answered in labs” typical of scientific research. While creationists may have religious motivations, motives alone don’t render work unscientific. Claims made by Young-Earth Creationism about the age of the Earth are, from a mainstream scientific standpoint, testable (and false). However, simply being testable isn’t sufficient for scientific status, as Popper himself noted.
The Challenge from History and Sociology
Thomas Kuhn’s historical approach to philosophy of science further complicated the picture. Kuhn argued that science operates largely within paradigms, which govern normal science. During periods of “normal science,” the paradigm is assumed correct, and empirical work focuses on solving puzzles within its framework, not testing the paradigm itself. Paradigm shifts occur during “scientific revolutions,” often following a crisis when anomalies repeatedly resist solution. Kuhn famously argued that rival paradigms can be incommensurable, meaning they lack a neutral, objective standard for comparison. While values like predictive accuracy and simplicity exist across paradigms, they are interpreted and weighed differently. For Kuhn, this historical reality suggested that science was less straightforwardly rational, cumulative, and progressive than previously thought.
Following Kuhn, the strong program in the sociology of science went even further, arguing that the acceptance of scientific theories is determined by social needs and interests, rather than simply evidence or rationality. They insisted on a symmetry principle, requiring explanations for accepted beliefs (even if we think them true and rational) to be of the same kind as explanations for rejected beliefs (even if we think them false and irrational) – typically, explanations appealing to local norms and non-epistemic interests. This view stands in stark contrast to traditional ideas of science as being epistemically special due to its methods or relationship with evidence.
Conclusion: No Simple Line
Given these challenges, most philosophers today agree that no single, adequate criterion for demarcation has been formulated It’s difficult to find a solid basis for distinguishing a poor scientific theory from a non-scientific one. If we give up on a strict demarcation criterion, it means we can’t simply declare something like creationism or astrology “unscientific” in that specific, fundamental sense.
Does this mean anything goes? Not necessarily. It might suggest that qualifying as “scientific” isn’t the ultimate goal; perhaps we should focus instead on whether a theory or practice is “good” science, based on criteria like predictive accuracy, explanatory power, consistency, and fruitfulness (values recognized even by Kuhn, though interpreted differently). The difference between science and pseudoscience might not be a sharp line, but rather a matter of degree, involving a complex combination of features like testability, empirical fit (while acknowledging tolerance for failures), progress compared to rivals, internal coherence, acceptance and critical scrutiny within relevant expert communities, and development of positive research programs rather than just negative arguments.
Ultimately, the quest to define what makes science intellectually special continues. While a simple, universally accepted definition of science distinguishing it from pseudoscience remains elusive, reflecting on the historical attempts and their limitations offers valuable insights into the complex nature of scientific knowledge itself.