1.

Albert Einstein proved that "space" and "time" have no meaning until you specify a reference body — then, by unifying two assumptions previously thought incompatible, he derived the Lorentz transformation, explained Mercury's orbit, and replaced Newton's inverse-square law with geometry.

2.

Physics deals in observable operations, not armchair intuitions

• Einstein insists a physical concept "does not exist for the physicist until he has the possibility of discovering whether or not it is fulfilled in an actual case" (§8)
• His definition of simultaneity — light signals from A and B meeting at midpoint M — is not a hypothesis about light but a "stipulation which I can make of my own freewill" to ground measurement in experiment (§8)
• "Every physical description resolves itself into a number of statements, each of which refers to the space-time coincidence of two events A and B" — all physics is about where and when things meet (§27)

3.

The principle of relativity and the constancy of light speed were thought incompatible — they weren't

• Special relativity's two postulates: (1) all Galileian coordinate systems are equivalent for natural laws; (2) light in vacuum travels at constant speed c = 300,000 km/sec regardless of source velocity (§13)
• The assumed contradiction arose from classical velocity addition: if light travels at c relative to the embankment, then relative to a train moving at v it travels at c − v — violating postulate 1 (§7)
• Einstein's resolution: "there is not the least incompatibility between the principle of relativity and the law of propagation of light" — the fault was in two tacit assumptions borrowed from classical mechanics (§7, §11)

4.

Classical mechanics smuggled in two unproven assumptions that had to go

• Assumption 1: "The time-interval (time) between two events is independent of the condition of motion of the body of reference" — dropped because simultaneity is relative, not absolute (§11)
• Assumption 2: "The space-interval (distance) between two points of a rigid body is independent of the condition of motion of the body of reference" — dropped because length contraction is real, not apparent (§11)
• Classical velocity addition W = v + w "cannot be maintained; in other words, the law that we have just written down does not hold in reality" (§6)

5.

The Lorentz transformation is the unique mathematical structure that preserves both postulates

• Einstein derives the transformation from light-signal equations and the principle of relativity — no ether, no prior hypotheses about matter structure (Appendix I)
• Length contraction: a metre-rod moving at v becomes √(1 − v²/c²) of a metre; time dilation: a moving clock ticks √(1 − v²/c²) seconds per second — at v = c the square-root becomes imaginary, marking c as "an unattainable limiting velocity" (§12)
• Fizeau's 1851 experiment on light in flowing water matched the relativistic formula within one per cent in Zeeman's repeat measurements — Lorentz had explained it with ad hoc hypotheses, but "the theory of relativity leads to the same law of motion, without requiring any special hypothesis whatsoever as to the structure and the behaviour of the electron" (§16)

6.

Mass and energy are not separate — they are the same thing

• A body absorbing energy E₀ increases its inertial mass by E₀/c² — the term mc² in E = mc² "is nothing else than the energy possessed by the body before it absorbed the energy E₀" (§15)
• "The law of the conservation of the mass of a system becomes identical with the law of the conservation of energy" — the two conservation laws, previously independent, are now one (§15)
• Einstein's caveat: in 1916, mass-changes from energy absorption were "not large enough to make themselves perceptible as a change in the inertial mass of the system" — the equation "has been thoroughly proved time and again since this time" (translator's note)

7.

The equivalence principle extends relativity to accelerated motion and gravity

• An observer in a chest pulled upward by a rope experiences identical forces to standing in a gravitational field on Earth — "his mode of grasping the situation violates neither reason nor known mechanical laws" (§20)
• This works because gravitational mass equals inertial mass for all bodies — "the same quality of a body manifests itself according to circumstances as 'inertia' or as 'weight'" (§19)
• A general principle of relativity: "All Gaussian co-ordinate systems are essentially equivalent for the formulation of the general laws of nature" — not just uniform motion, but any motion (§28)

8.

Gravity is not a force — it is the geometry of space-time, and matter curves it

• On a rotating disc, a measuring-rod shrinks tangentially but not radially, so circumference divided by diameter is no longer π — "the propositions of Euclidean geometry cannot hold exactly on the rotating disc, nor in general in a gravitational field" (§23)
• The marble-table analogy: heat the centre, rods expand, squares break down — "just as it was there impossible to construct a Cartesian coordinate system from equal rods, so here it is impossible to build up a system from rigid bodies and clocks" (§27)
• Light rays in a gravitational field are no longer straight lines — for grazing incidence past the Sun, the predicted deflection is 1.7 seconds of arc, half Newtonian attraction and half space-time curvature (§22)

9.

The general theory explains three phenomena that broke classical mechanics

• Mercury's perihelion precession: Leverrier in 1859 and Newcomb in 1895 found an unexplained 43 seconds of arc per century; general relativity calculates exactly 43 seconds — "strictly in agreement with observation" (§29)
• Light deflection during a solar eclipse: Eddington's 1919 expeditions to Sobral (Brazil) and Principe (West Africa) photographed stars near the eclipsed Sun, confirming the theory; the discrepancy expected between eclipse and comparison photographs was "a few hundredths of a millimetre only" (Appendix III)
• Gravitational redshift: Adams confirmed the effect on the dense companion of Sirius in 1924, where the shift is roughly thirty times greater than for the Sun (translator's note, Appendix III)

10.

Newton's inverse-square law emerges as a first approximation — not a fundamental postulate

• In weak fields and low velocities, general relativity "then obtains as a first approximation the Newtonian theory" — Newton's hypothesis that gravity decreases as inverse square "is here obtained without any particular assumption" (§29)
• Classical mechanics required the universe to be a "finite island in the infinite ocean of space" to avoid infinite fields; the general theory connects the radius of the universe R to average matter density ρ via R = √(1/(2πκρ)) — a finite, unbounded spherical space is the natural geometry (§§31–32)
• Appendix IV (added at the fifteenth reprinting) acknowledges that "a strange difficulty" remains: Hubble's expansion implies an origin "only about 10⁹ years ago, while physical astronomy makes it appear likely that the development of individual stars and systems of stars takes considerably longer"

11.

Einstein's "two postulates" are not two independent foundations — and he admits it

• The constancy of light speed is not a second postulate alongside relativity; it is itself derived from Lorentz's electrodynamics, which Einstein calls "a necessary consequence" of "epoch-making theoretical investigations" (§7) — the theory's empirical base is one assumption dressed in two hats
• Einstein himself says he "purposely treated the empirical physical foundations of the theory in a 'step-motherly' fashion" (Preface), then acknowledges Lorentz had already explained Fizeau and electron motion with electrodynamics alone — the theory's "heuristic value" is simplification, not discovery of new data (§§13–14, §16)
• This is not a deduction from first principles but a re-axiomatisation of pre-existing physics — the Lorentz transformation, contraction, and time dilation were all known before 1905, and Einstein's contribution was to show they follow from fewer assumptions, not to prove them true

12.

Unifying axioms is the highest form of empirical discovery

• Both Einstein's defenders and his critics share an unexamined assumption: that a theory's value is measured by how many new predictions it makes — but Einstein himself says "the 'truth' of the theory lies in the fact that it correlates a large number of single observations" (Appendix III)
• Maxwell unified electricity and magnetism; Darwin unified embryology and biogeography; Einstein unified space, time, gravity, matter, and energy into a single geometric manifold — each time, the predictive content was largely pre-existing, and the revolution was in showing that once separate phenomena were actually the same phenomenon seen from different angles
• The 43 arcseconds of Mercury's perihelion was not a prediction anyone asked for; it was a discrepancy Newtonian mechanics had papered over with ad hoc hypotheses for sixty years — general relativity's greatest achievement was not predicting something new but making an old anomaly mathematically inevitable, replacing a patch with a proof

1.