What separates science from pseudoscience, and is falsifiability the right criterion?
Explain Popper's falsificationism as a solution to the demarcation problem and assess its strengths and weaknesses, including the Duhem-Quine challenge
A focused answer on Popper's falsificationism. The demarcation problem, why Popper rejects verification for falsifiability, conjecture and refutation, corroboration, and the main objections including the Duhem-Quine problem and naive versus sophisticated falsificationism.
Reviewed by: AI editorial process; not yet individually human-reviewed
Have a quick question? Jump to the Q&A page
Jump to a section
What this dot point is asking
SEAB wants you to explain Karl Popper's falsificationism and to assess it as an answer to the demarcation problem, the problem of distinguishing genuine science from non-science or pseudoscience. Popper's view is one of the most influential in the philosophy of science, and it responds directly to the problem of induction. Your task is to set out the criterion, the logic behind it, and the serious objections it faces.
The answer
The demarcation problem
The demarcation problem asks what distinguishes scientific theories from non-scientific ones. Popper was struck that some theories he examined could explain any conceivable observation: whatever happened, a confirming interpretation was available. He contrasted these with theories that made bold, specific predictions which could have turned out wrong. This contrast suggested that testability, not explanatory flexibility, is the mark of science.
Falsifiability as the criterion
Popper's answer is that a theory is scientific if and only if it is falsifiable: it forbids some observable state of affairs, so that some possible observation would refute it. A theory that is compatible with every possible observation forbids nothing and so is unfalsifiable and unscientific. Crucially, falsifiability is about logical form, what the theory rules out, not about whether the theory is true or has actually been refuted.
Why verification fails and falsification works
Popper rejects verification (accumulating confirming instances) for two reasons. First, the problem of induction: no number of confirming instances can justify a universal law. Second, the logical asymmetry between verification and falsification. A universal law of the form "all A are B" cannot be proved by any finite number of observed A that are B, because the next A might differ. But a single A that is not B refutes the law with deductive certainty, by modus tollens. So while laws cannot be verified, they can be refuted, and Popper makes refutation the engine of science.
Conjecture, refutation and corroboration
The scientific method, on this view, is conjecture and refutation. Scientists propose bold, falsifiable hypotheses and then try as hard as they can to refute them by severe tests. A hypothesis that survives serious attempts at refutation is corroborated, but never proved or even rendered probable; corroboration is a record of past testing, not a measure of likely truth. Science advances by eliminating false theories, not by confirming true ones.
The main objections
Three objections bite. The Duhem-Quine problem: a hypothesis is never tested in isolation but always with auxiliary assumptions, so a failed prediction does not tell us whether the core hypothesis or an auxiliary is false; clean falsification is therefore impossible. Second, much legitimate science is hard to falsify, including theories in their early development, probabilistic claims, and historical sciences; a sharp falsifiability line risks excluding them. Third, as Kuhn observed, scientists rightly do not abandon a well-established theory at the first anomaly, retaining it until a better alternative appears, which conflicts with naive falsificationism. These objections motivate a sophisticated falsificationism that judges whole research programmes over time rather than single refutations.
Examples in context
Example 1. A risky prediction. A theory that predicts light from distant stars will bend by a specific, measurable amount when passing the sun makes a bold, falsifiable claim: had the measured deflection differed, the theory would have been refuted. Popper treated such risky, refutable predictions as the hallmark of good science, contrasting them with theories that can absorb any result. The case shows why testability, not mere explanatory power, is his criterion.
Example 2. The unfalsifiable defence. Suppose a forecaster predicts a market crash, and when none comes, says the crash is merely delayed, and when the market rises, says this is the calm before the storm. No outcome is allowed to count against the prediction, so it forbids nothing and is unfalsifiable. The example illustrates the kind of any-outcome-fits reasoning that Popper's criterion is designed to exclude from science.
Try this
Q1. State Popper's criterion of demarcation and explain what "falsifiable" means. [6 marks]
- Cue. A theory is scientific if and only if it is falsifiable, that is, it forbids some observable state of affairs so that a possible observation could refute it; falsifiability is about logical form, not truth.
Q2. Explain why Popper prefers falsification to verification. [8 marks]
- Cue. Verification fails because of the problem of induction and the asymmetry that no instances can prove a universal law while one counterexample refutes it; so science should seek refutations, not confirmations.
Q3. Explain the Duhem-Quine objection to naive falsificationism. [6 marks]
- Cue. A hypothesis is always tested with auxiliary assumptions, so a failed prediction does not show whether the core hypothesis or an auxiliary is false; clean falsification of a single hypothesis is therefore impossible.
Exam-style practice questions
Practice questions written in the style of SEAB exam questions on this dot point, with worked answer explainers. The year tag is the paper they imitate, not the source.
Original20 marksIs falsifiability a satisfactory criterion for distinguishing science from non-science? Discuss.Show worked answer →
A strong answer explains the demarcation problem and Popper's solution. A theory is scientific if it is falsifiable: it forbids some observable state of affairs, so that some possible observation would refute it. Verification fails as a criterion because confirming instances are cheap and induction is unjustified; falsification exploits the logical asymmetry that a single counter-instance can refute a universal claim (modus tollens) though no number of instances can prove it.
Give Popper's contrast cases: Einstein's theory made a risky, falsifiable prediction about starlight bending; some theories, as Popper saw them, can accommodate any observation and so forbid nothing, which he treated as a mark of unfalsifiability.
Evaluate. Strengths: it captures the boldness and testability of good science, explains the value of risky predictions, and avoids the problem of induction. Weaknesses: the Duhem-Quine problem, that a failed test never refutes a single hypothesis in isolation because auxiliary assumptions are always involved, so clean falsification is impossible; the existence of legitimate but hard-to-falsify science (early stages of a theory, probabilistic and historical claims); and the fact that scientists rightly retain refuted theories pending a better alternative (Kuhn's point).
Judgement: falsifiability captures something important about scientific spirit but fails as a sharp, sufficient criterion; sophisticated falsificationism softens it. Markers reward the demarcation problem, the asymmetry argument, contrast cases, the Duhem-Quine objection, and a decided verdict.
Original12 marksExplain the logical asymmetry between verification and falsification that motivates Popper's view.Show worked answer →
The expected answer states the asymmetry in terms of universal generalisations. A scientific law typically has the form "all A are B." No finite number of observed A that are B can prove this, because the next A might not be B; so universal laws cannot be verified by instances (and induction, which would bridge the gap, is unjustified).
However, a single observed A that is not B refutes "all A are B" with deductive certainty, by modus tollens: if the law implies this A is B, and it is not, the law is false. So one counterexample can falsify a universal claim even though no number of confirming cases can verify it.
Draw the moral Popper draws: since theories cannot be proved but can be refuted, the rational scientific strategy is to propose bold, falsifiable conjectures and try hard to refute them, retaining those that survive as corroborated (not proved). Note the limit: this asymmetry is clean only if the auxiliary assumptions used to derive the prediction are not themselves in question (the Duhem-Quine caveat).
Judgement-style close: the asymmetry is logically genuine and motivates falsificationism, but real refutation is messier than the bare logic suggests. Markers reward the universal-law form, the modus tollens point, the conjecture-and-refutation moral, and the auxiliary-assumption caveat.
Related dot points
- Explain Hume's problem of induction and the new riddle of induction, and assess the main responses including pragmatic, probabilistic and Popperian replies
A focused answer on the problem of induction. Hume's argument that inductive inference cannot be justified non-circularly, Goodman's new riddle, and the main responses: pragmatic vindication, probabilistic accounts, and Popper's rejection of induction.
- Characterise the scientific method, contrasting inductivist and hypothetico-deductive accounts, and assess whether a single method defines science
A focused answer on the scientific method. The inductivist picture, the hypothetico-deductive model, the roles of observation, hypothesis, prediction and testing, and whether any single method captures what makes inquiry scientific.
- Explain Kuhn's account of paradigms, normal and revolutionary science, and incommensurability, and assess what it implies for scientific objectivity and progress
A focused answer on Kuhn's philosophy of science. Paradigms, normal science and puzzle-solving, anomalies and crisis, revolutionary paradigm shifts, incommensurability, and the challenge this poses to a cumulative, fully objective picture of science.
- Contrast scientific realism with instrumentalism and anti-realism, and assess the no-miracles argument and the pessimistic meta-induction
A focused answer on the realism debate in science. Whether theories about unobservables are true descriptions or mere predictive instruments, the no-miracles argument for realism, the pessimistic meta-induction against it, and structural and constructive empiricist middle positions.