Einstein’s 1905 derivation unified mass and energy conceptually — not empirically. It replaced two conservation laws with one. It did not predict applications. It did not contain E = mc² as written. Its power lies in logical necessity, not experimental novelty.
On the Origin of Species (1859) introduced natural selection as the mechanism for evolutionary descent with modification. It presented evidence for common descent via a branching pattern, drawn from Darwin’s Beagle expedition, research, correspondence, and experimentation. It established descent with modification as the dominant scientific explanation of natural diversification. Its proposition of universal common ancestry is now accepted—but that acceptance postdates the book and is not itself what the book established.
Schrödinger's cat is a 1935 thought experiment designed by Erwin Schrödinger in correspondence with Einstein to critique the Copenhagen interpretation of quantum mechanics. It uses a sealed chamber containing a cat, radioactive source, Geiger counter, relay, hammer, and flask of hydrocyanic acid. The setup links a single quantum event — atomic decay — to a macroscopic outcome. Mathematically, the wave function describes a superposition of 'alive' and 'dead', yet observation always reveals one definite state. Schrödinger intended this as reductio ad absurdum, not as a proposal. No experiment was conducted. The work establishes a tension between formalism and observation — nothing more.
The 1998 supernova measurements established cosmic acceleration via distance–redshift mismatch — not theory, not analogy, but empirical inference from standard candles. It overturned the assumption of gravitational deceleration. It did not identify dark energy. It did not prove Λ. It remains the strongest direct evidence for accelerated expansion — and the clearest signal of our ignorance about 70% of reality.
Penzias and Wilson discovered a 3.5 K isotropic microwave background in 1964 using the Holmdel Horn Antenna at Bell Labs. They eliminated local interference — including pigeon droppings — and confirmed the signal originated outside the Milky Way. Its match to Dicke’s prediction prompted interpretation as the Big Bang’s remnant. But the work itself claimed only a persistent, unexplained antenna temperature — not cosmology.
The Double Helix is not the discovery of DNA’s structure — it is Watson’s memoir of it. The real discovery is in the 1953 Nature papers. The book’s value lies in exposing how scientific credit gets assigned, not in explaining how the structure was found.
Young’s 1801 double-slit experiment established wave behaviour for visible light. Later versions with electrons (1927) extended it to matter. The interference pattern is real. The disappearance of that pattern upon path detection is real. But Young did not observe particles, nor did he claim duality — he argued for waves only. Duality emerged from later experiments combining his method with quantum detection.
The Asch experiments show that unanimous group pressure causes public conformity on simple perceptual judgments — not because people misperceive, but because they yield in speech. The effect is real, quantified, and limited: it depends on unanimity, not large numbers; it was tested only on white male undergraduates; and it measures response, not belief or vision.
Bayes's 1763 essay solves a narrow inverse probability problem using conditional probability. It computes limits on a binomial parameter—not a general theory of belief updating. Its power lies in precision, not generality.
BDNF was isolated from pig brain tissue in 1982 by Yves-Alain Barde and Hans Thoenen, identifying it as a distinct protein.
Chicago Pile-1 proved a self-sustaining nuclear chain reaction was possible using natural uranium and purified graphite — nothing more, nothing less.
Geoffrey Hinton and colleagues established core methods—backpropagation for representation learning, and Boltzmann machines for unsupervised deep generative modelling—between 1985 and 1995. Their work was grounded in statistical mechanics, experimentally verified on small-scale networks, and explicitly recognised as foundational for machine learning with artificial neural networks. It did not deliver production AI, nor claim to.
