What Makes Something "Scientific"? The Demarcation Problem

Scientific Methodology

Chapter 1 · What Makes Something "Scientific"? The Demarcation Problem

Not every claim that sounds authoritative, uses technical language, or comes from a genuine expert is actually scientific in a meaningful sense. This course covers the real methodology underneath science itself — starting with the real observation that first made philosopher Karl Popper ask what separates science from everything that merely resembles it.

A Real Comparison That Sparked a Question

In autumn 1919, Popper — then a young man in Vienna, and briefly a former Marxist — noticed something genuinely striking: Albert Einstein's general theory of relativity and the psychological theories of Sigmund Freud and Alfred Adler were both widely treated as serious, authoritative bodies of knowledge. But they behaved completely differently when tested against the real world.

A Real, Risky Prediction — Eddington's 1919 Eclipse Expedition

Einstein's theory made a genuinely risky, specific prediction: light from distant stars should bend by a precise, calculable amount as it passed near the sun's gravity. A British expedition led by Arthur Eddington tested this directly during the solar eclipse of 29 May 1919 — and the theory could have failed. It didn't; the measured deflection matched Einstein's prediction.

Freud's and Adler's theories, by contrast, seemed able to explain literally any human behavior after the fact — Popper later wrote there was "no conceivable human behaviour which could contradict them." Whatever someone did could always be reinterpreted as confirming the theory. Marxism, in Popper's own real assessment, had drifted into the same pattern.

The Falsifiability Criterion

Popper formalized this real distinction in his 1934 book Logik der Forschung (translated into English in 1959 as The Logic of Scientific Discovery): a genuinely scientific theory must make predictions specific enough that they could, in principle, be proven wrong by a real observation. It's not a requirement that a theory actually be false — general relativity passed its real test — only that a real test capable of failing it must exist at all.

A Real, Concrete Example: Astrology

Popper himself pointed to astrology as a real case of the problem: horoscope predictions are typically vague enough to fit almost any real outcome after the fact, with no clear, specific claim that a contradicting event could ever actually falsify.

Falsifiable vs. Unfalsifiable, Compared

General Relativity (1919)	Psychoanalysis (Popper's real assessment)
Made a specific, risky, real prediction	Could accommodate any real behavior after the fact
A real test (the 1919 eclipse) could have failed it	No real, conceivable observation could contradict it
Passed a genuine, dangerous test	Never faced one

The Real, Practical Lesson

Falsifiability is a test of a theory's own structure, not a verdict on whether it happens to be true. Asking "what real observation would prove this wrong?" is often more revealing than asking "what evidence supports this?" — since a theory built to accommodate any possible evidence supports itself automatically, which is precisely the real problem.

What This Course Covers

Theories & Laws

Real definitional distinctions

Experiment Design

Variables, control groups, randomization

Blinding

Why the placebo effect demands it

Paradigm Shifts

How consensus actually changes

Reproducibility

The real, ongoing crisis

Pseudoscience

Demarcation applied to real cases

Open Science

Preprints and publishing reform

Self-Correction

Retraction and real fixes

Capstone

Design and critique a real experiment

Hands-On Exercises

Exercise 1

Explain, in your own words, exactly what real observation from the 1919 eclipse expedition would have falsified Einstein's theory — what specifically did Eddington's team measure, and what result would have proven the prediction wrong?

📄 View solution

Exercise 2

Take a real claim you've heard recently (not necessarily scientific) and ask Popper's own question of it: what specific, real observation would need to occur to prove it false? If you genuinely can't think of one, explain what that tells you about the claim.

📄 View solution

Exercise 3

Explain, in your own words, why Popper's criterion evaluates a theory's own falsifiability rather than whether it happens to be true — using general relativity's own real 1919 test (which the theory passed) as the example.

📄 View solution

Chapter 1 Quick Reference

Karl Popper's real 1919 observation contrasted Einstein's risky, testable general relativity against psychoanalysis and Marxism's inability to be contradicted by any real observation
Arthur Eddington's real 1919 solar eclipse expedition tested Einstein's theory directly — and it could have failed
Popper formalized falsifiability in his real 1934 book, Logik der Forschung
A genuinely scientific theory must permit a real observation capable of proving it wrong, even if that observation never actually occurs
Popper himself cited astrology as a real example of claims too vague to be falsifiable