Emmy Noether studied formulas that give every possible trip a number, and so pick out the trip nature takes: the arc of a thrown ball, the swing of a spring, the quickest slide from A to B.
Change the picture and ask the formula again: start the clock later, move everything one step, turn it. If the formula gives the same number, the change doesn't matter to it. And every change that doesn't matter hands you a prize: a number that stays the same all along the trip. Energy is one of them.
Watch a change. Say whether it matters.
Every change that doesn't matter wins a prize.
Nine changes, one answer each. Right answers earn badges: red for the paper's lessons, pink for its traps.
Watch a change. Say whether it matters.
Every change that doesn't matter wins a prize.
Tap the picture to watch the trip again, as often as you like. Then give one answer: does the change matter to the formula?
Noether's paper is about formulas that give every possible trip a number, and so pick out the trip nature takes. She proved two things.
First: whenever the formula doesn't care about a change, such as moving the picture, turning it or starting the clock later, there is a number that stays the same all along nature's trip: a law, which a fake trip can break. That is where energy, momentum (the sideways speed) and spin come from. It works backwards too: every such number comes from a change the formula doesn't care about. And you can read it off the formula itself, as the cards showed: no t in it, no energy lost.
Second: some formulas don't care about a much bigger change, one you can make differently at every moment, such as changing how fast the clock runs. Such a change wins no prize but a rule: the equations of motion are not independent, as the slide's two equations fix only its shape. And since starting the clock later is just one such change, the energy it should give comes out zero on every trip, real or fake, so it tells you nothing.
Einstein's theory of gravity is of this second kind. Its formula, like the slide's, doesn't care how fast the clocks run, differently at every place and moment. So its energy law comes out zero whatever happens, real or fake, and says nothing.
Hilbert had claimed that Einstein's gravity has no real law of energy. Noether showed why: its formula doesn't care how fast the clock runs, so its energy stays the same on every trip, real or fake. A law that even a fake can't break is no law.
Watch a change. Say whether it matters.
Every change that doesn't matter wins a prize.
The formula. Each picture has a formula, in symbols and in words. It gives every possible trip a number, S, and so picks out the trip nature takes. x and y say where the dot is (x across, y up); x′ and y′ say how fast they change; ∫ … dt means: add up along the trip. The dots on a trail are the ticks of a clock: far apart where the dot is fast, close together where it is slow.
A card. Each of the nine cards shows a trip, which you can watch again as often as you like by tapping it, and a change: start the clock later, move everything one step to the left, turn everything round the peg, or change the timing so the dot speeds up here and slows down there. You give one answer: it matters (the formula gives a different number), or it doesn't (the same number).
The answer. A second picture appears below the first: the trip after the change. Tap either picture to run both again. The second camera moves or turns with the change, so if the change doesn't matter, the two pictures look the same. Below them: the formula's number before and after, and the part of the formula that decides the answer. Every answer can be read off the formula.
The prizes. A change that doesn't matter wins a prize: a number that stays the same all along the trip. A graph shows it next to things that do change, such as speed and height. Some cards also show a fake: a trip nature never makes. A real law stays the same along the real trip, and a fake can break it.
The badges. Every right answer earns a badge. Red badges are the paper's lessons: energy, sideways speed and spin; energy as a real law, which a fake breaks; the slide's energy, which comes out zero; and the rule the slide gets instead. Pink badges are the traps: a change that does matter, so there is no prize. A wrong answer leaves its badge grey.
The end. Your shelf, a sentence about it, and what the paper says.
Nine changes to small pictures, each with its formula on screen: say whether the change matters to the formula. Every change that doesn't matter wins a prize, a number that stays the same along the real trip. The ball's formula cares how fast the clock runs, so a fake throw breaks its energy: a real law. The slide's formula doesn't, so its energy is zero on every slide, real or fake: no law. Einstein's gravity is like the slide, as Hilbert had claimed and Noether proved.
Emmy Noether's 1918 paper is about formulas of the kind S = ∫ f dt, which give every trip a number and so pick out the trip nature takes (mathematicians call them variational integrals), and about the changes that leave such a formula unchanged. Theorem I: for every change of a fixed kind that leaves the formula unchanged (start the clock later, move, turn; a group with a fixed number of parameters) there is a quantity that stays the same along every real motion, a conservation law; and the other way round. Theorem II: if the formula doesn't care about a change that can be made differently at every moment (a group that depends on free functions), the equations of motion are not independent: some of them follow from the others; and the other way round. Her first example of Theorem II is a formula that reads only the shape of a curve, not how fast it is traced, with the identity Σψᵢuᵢ′ = 0. Section 6 answers Hilbert: when starting the clock later is one of these free changes, its conservation law is improper: an identity that holds on every path, whether or not the equations of motion do. Proper energy laws exist exactly when there is no such free change. Hilbert's claim that this failure is characteristic of general relativity holds, she ends, when 'general relativity' means a formula that doesn't care about such free changes.
'Does the change matter?' is her invariance condition (equation 11). A prize is a conserved quantity (equations 12 and 13). Cards 1, 3 and 5 are Theorem I: energy, momentum (the sideways speed) and spin. Cards 7, 8 and 9 are section 6 and Theorem II. Card 7 shows the ball's formula caring how fast the clock runs, and a fake (the re-timed throw) breaking the ball's energy: a proper law. Card 8 shows the slide's energy improper, zero on the real slide and on a straight one alike. Card 9 shows the slide's formula indifferent to any change of timing, with her identity as the rule it gets instead of a prize. The pink cards (growing gravity, the magnet, the wind) are Theorem I the other way round: where the formula cares, there is no law. Each 'reason, in the formula' line names the part of the formula that decides the card. The formula is on screen throughout, because the formula is the paper's subject.
This game demonstrates the theorems on six small formulas, computed, not quoted. Every card's answer is her invariance condition, checked on random paths, and for the change of timing on random timings too. Every prize is her conserved quantity, computed along real motions (paths that obey the equations of motion). The slide's energy is zero on every path, improper; the re-timed slide satisfies her identity Σψᵢuᵢ′ = 0; the ball's energy wobbles along the re-timed throw; the real slide is quicker than the straight one and than hundreds of bent ones; and every sentence the game gives as a reason is checked against the formula, point by point. The formula's numbers and the two pictures agree with the computed answer on every card.
It assumes the six worlds: a thrown ball; the same ball under gravity that grows with time; the ball with a magnet; a dot on a spring round a peg; the same in a wind; and the slide, whose formula (each bit of the way divided by the speed there: the time) is the classic quickest-slide problem, the brachistochrone. The changes shown (five minutes later, one step left, a small turn, a change of timing) are finite versions of the paper's infinitely small ones. The straight comparison slide, the threshold for 'the same number', and all the worlds' numbers are ours, in the config block at the top of the source, with their reasons.
The paper establishes the two theorems and their converses for any number of variables and derivatives, and the section 6 result on proper and improper energy laws. It does not say what energy 'is' in general relativity, and neither does the game; the slide is a small picture of the kind of freedom Einstein's theory has, not a model of gravity.
Formulas with higher derivatives and with several variables of integration (fields); mixed groups; how finite changes are built from infinitely small ones; section 5, on relative invariance; Einstein's modified formula and the exchange with Klein in the notes; the group structure itself.
Emmy Noether, Invariante Variationsprobleme. Nachrichten von der Königlichen Gesellschaft der Wissenschaften zu Göttingen, Mathematisch-physikalische Klasse, 1918, pp. 235–257. No DOI. German text at de.wikisource.org/wiki/Invariante_Variationsprobleme; English translation by M. A. Tavel, Transport Theory and Statistical Physics 1 (1971) 186–207, arXiv:physics/0503066.
Game built from this paper; the paper is the authority where the two disagree. Fonts: Chakra Petch, Saira and Share Tech Mono (SIL Open Font License).