The Equivalence Principle and Curved Spacetime
Starts from the puzzle of Newtonian gravity in special relativity and the equality of inertial and gravitational mass, tested from Galileo to the MICROSCOPE satellite, and states the equivalence principle. From the principle alone follow gravitational time dilation, measured by Pound and Rebka over 22.5 metres and corrected for every day in the clocks of the GPS, and the bending of light, and the unit ends with gravity as the curvature of spacetime, the picture the full theory makes precise.
Seven units have taken the two postulates of 1905 as far as they go, and gravity has stayed outside them. That is no accident. Newton’s law of gravitation and special relativity cannot both be right as they stand, and Albert Einstein (1879-1955) needed ten years, from 1905 to 1915, to replace the one with a theory that contains the other. The first step of those ten years is the subject of this unit. It is an idea so simple that it fits in one sentence, and from that sentence alone follow three predictions that have since been measured on towers, in aircraft, in satellites, and across a single millimetre of cold atoms.
1 Gravity and the two masses
Newton’s law says that two bodies of masses \(M\) and \(m\) a distance \(r\) apart attract each other with the force \(GMm/r^2\), where \(G = 6.67 \times 10^{-11}\unit{m^{3}\,kg^{-1}\,s^{-2}}\) is the gravitational constant of Unit 1. The law contains no time, so the force would have to act instantly, however far apart the two bodies are. If the Sun were somehow removed, the Earth would leave its orbit at that same instant, although the last sunlight would still be on its way for 8.3 minutes. Special relativity allows no signal faster than light, and Unit 5 showed why a faster one would let an effect precede its cause, so a force that acts instantly across 150 million kilometres cannot be the whole story. The second trouble is less obvious. The laws of Units 2 to 7 hold in inertial frames, and Unit 1 defined an inertial frame as one in which a body left alone moves in a straight line at constant speed. In a room on the Earth nothing left alone does that; everything falls. Newton’s answer was to call the room inertial and to treat gravity as a force like any other. Einstein’s answer, in Section 2, will be the opposite one, and it is possible only because of an experimental fact that Newton knew and could not explain.
1.1 Inertial mass and gravitational mass
The symbol \(m\) does two different jobs in Newtonian mechanics. In the second law, \(\vec{F} = m\vec{a}\), it measures how much a body resists being accelerated, and this is called the inertial mass \(m_{\mathrm{i}}\). In the law of gravitation it measures how strongly the body is pulled, and this is the gravitational mass \(m_{\mathrm{g}}\). There is no reason in the laws themselves why these two should be the same; electric charge, for example, is what a field pulls on, and it has nothing to do with inertia. A body falling near the Earth obeys \(m_{\mathrm{i}}\,a = m_{\mathrm{g}}\,g\), so its acceleration is \[ a = \frac{m_{\mathrm{g}}}{m_{\mathrm{i}}}\,g , \] and all bodies fall at the same rate only if the ratio \(m_{\mathrm{g}}/m_{\mathrm{i}}\) is the same for all of them. That they do fall at the same rate, once the air is removed, is the universality of free fall, and it is one of the best-tested facts in physics.
The test has been refined for four centuries, and Table 1 lists its stages. Galileo Galilei (1564-1642) compared pendulums of different materials; the story of the Leaning Tower of Pisa was told by his pupil Viviani after his death and is probably a legend. Isaac Newton (1643-1727) swung pendulums of gold, silver, lead, glass, sand, salt, wood, water, and wheat and found their periods equal to one part in a thousand. Then came the torsion balance of the excursion, and in 1964 the method of Robert Dicke (1916-1997), who compared the fall of aluminium and gold towards the Sun rather than towards the Earth. Because the laboratory turns with the Earth, the direction to the Sun swings round once a day, so any difference between the two materials would have reversed its sign every twelve hours and could have been separated from every disturbance that did not. The modern measure of how well the two masses agree is the Eötvös parameter, the difference of the accelerations of two bodies \(A\) and \(B\) in the same field, divided by their mean: \[ \eta = \frac{2\,(a_A - a_B)}{a_A + a_B} . \] A nonzero \(\eta\) for any pair of materials would mean that the two masses differ, and Table 1 shows how far the search has been pushed; each entry is an upper limit, the largest value of \(\eta\) that the experiment could have missed.