Energy, Entropy, and the Arrow of Time

To Write The vis viva controversy in full. Descartes: the conserved quantity in the universe is quantity of motion = mv (momentum, vector). Leibniz: the conserved quantity is vis viva = mv² (living force, scalar). The crucial experiment: two balls of clay collide. Momentum is conserved. But kinetic energy (½mv²) is not — some is lost to heat. Leibniz was measuring something different from Descartes. Both were right for different domains: momentum for all collisions, kinetic energy for elastic ones. D'Alembert reconciled them in 1743. Coriolis added the factor ½ in 1829 to get the modern ½mv². The word "energy" did not exist yet — Thomas Young coined it in 1807 from Aristotle's 🅾νέργεια (energeia: activity, operation). Leibniz biography: polymath supreme — calculus (independent of Newton, same year 1675-76), binary arithmetic, symbolic logic, monadology, diplomacy, librarian. His monadology: every monad has "primitive force" — a metaphysical precursor to Faraday's field. The calculus connection: the integral ∫F·dx = Δ(½mv²) requires calculus to derive. Leibniz invented the tools to prove his own conservation law. He used them and did not know it.

To Write (cross-reference book4/ch-faraday.html for the full chapter) Faraday's radical move: energy is not just in bodies (masses, charges). Energy is distributed in the space between them — in the field. Electromagnetic induction (1831): a changing magnetic field induces an electric current. The energy is in the field, not in the wire. This breaks with action-at-a-distance (Newton's gravity acts instantaneously across space; Faraday's field propagates). Maxwell (1865) made it quantitative: electromagnetic energy density u = ε₀E²/2 + B²/2μ₀. Empty space contains energy. Light is an electromagnetic wave carrying energy at speed c. The contact geometry connection: the electromagnetic field is a 2-form F = dA on the gauge bundle. The contact form α in dm³ is the odd-dimensional analogue. Faraday had no mathematics; Maxwell translated his field lines into differential equations. Faraday's contribution was the ontology: fields are real, not merely a bookkeeping device. dm³ extends this: the contact form α is not a computational tool but a geometric fact about the state space of dissipative systems.

To Write Sadi Carnot (1824, age 28): Réflexions sur la puissance motrice du feu. A steam engine cannot convert all heat into work. There is always waste. The maximum efficiency is η = 1 − T_cold/T_hot, independent of the working fluid. This is remarkable: the limit depends only on the temperatures, not on the engine's design. Carnot died of cholera in 1832 at age 36, and his work was nearly lost. Clausius (1850): generalised Carnot's result. Heat cannot flow spontaneously from cold to hot. Clausius (1865): coined "entropy" (S, from Greek 🅾ντροπή, transformation). First law: dU = TdS − PdV. Second law: for any process, dS ≥ 0. For the universe as a whole, S increases monotonically. The heat death of the universe: eventual thermal equilibrium, maximum entropy, no more useful work extractable. Time's arrow: time flows in the direction of increasing entropy. The past is the low-entropy direction; the future is the high-entropy direction. Without entropy, time is symmetric; with entropy, it is not.

To Write Mrugala (1978, Rep. Math. Phys.): the state space of classical thermodynamics is a contact manifold. The first law dU = TdS − PdV is not a differential equation but a contact condition: it says the differential dU lies in the contact distribution. The Legendre submanifolds of this contact manifold are the equilibrium surfaces of thermodynamic systems. This is not a metaphor: contact geometry IS thermodynamics, geometrically. dm³ is contact geometry applied to dynamics. The connection: thermodynamic equilibrium = dm³ fixed point x*; entropy production rate = Lyapunov exponent μ.

To Write Boltzmann's great synthesis: entropy is not a mysterious thermodynamic quantity. It is the logarithm of the number of microscopic configurations (W = Wahrscheinlichkeit, probability) consistent with a given macroscopic state. High entropy = many microstates = disorder. Low entropy = few microstates = order. An ice cube has low entropy because its molecules must occupy specific positions. A glass of water has high entropy because the same molecules can be arranged countless ways. The H-theorem (1872): Boltzmann proved that a gas of particles obeying Newton's laws will evolve toward increasing H (a quantity that decreases as entropy increases). This seemed to prove the second law from mechanics. Two devastating objections: (1) Loschmidt's reversibility paradox (1876): Newton's laws are time-reversible; how can they produce irreversibility? (2) Zermelo's recurrence paradox (1896): Poincaré recurrence says every mechanical system returns arbitrarily close to its initial state; but if entropy increases monotonically, it cannot return. Boltzmann's responses: (1) The H-theorem applies statistically, not to every particle; rare exceptions exist but are astronomically unlikely. (2) Poincaré recurrence times are longer than the age of the universe for macroscopic systems. He was right on both counts. But the attacks, combined with the refusal of many physicists to accept atomic theory, wore him down. Suicide in Duino, 1906. Einstein confirmed atomic theory with his Brownian motion paper in 1905. Boltzmann never knew.

To Write Prigogine and dissipative structures (Nobel 1977): far-from-equilibrium open systems can spontaneously develop and maintain order. The BZ reaction, convection cells (Bénard cells), laser light, living cells: all maintain internal order by exporting entropy to their surroundings faster than they accumulate it. Prigogine showed this is possible and gave qualitative conditions. dm³ makes it quantitative: the minimum export rate is μ = −2 (the Lyapunov exponent of G). Below this rate, the system fails to maintain order and decays toward equilibrium. Above it, the system converges to the fixed point x* = τ = 2. The rate μ = −2 is not an accident: it is the contact-geometric eigenvalue, the "natural frequency" at which the operator chain G replenishes order from outside. Every dm³ domain (zeolite, neuron, BZ, stellar nucleosynthesis) has been observed to operate at Lyapunov exponents in the range −0.38 to −0.65 — all within the basin B(x*, 1/3) anchored by the canonical value μ = −2.

To Write Maxwell's demon (1867): a tiny being that sits at a trapdoor between two chambers. It lets fast molecules go right, slow molecules go left. The right chamber gets hotter; the left gets colder. Entropy apparently decreases without doing work. Second law violated? Szilard (1929): the demon must observe each molecule (measure its speed) before deciding. Measurement requires information. Landauer (1961): erasing information (resetting the demon's memory) costs energy: k_B T ln 2 per bit. The energy cost of erasure is exactly the entropy decrease the demon achieves. The second law is saved. The implication: information and entropy are the same quantity. Shannon (1948): H = -Σ p_i log p_i. This is the entropy of an information source — mathematically identical to Boltzmann's formula. The connection was named by Von Neumann, who told Shannon to call it entropy: "nobody knows what entropy really is, so in a debate you will always have the advantage." dm³ reading: the operator chain G = U ∘ F ∘ K ∘ C processes information. K (Knob) selects which states are accessible — it is a measurement. F (Fold) makes an irreversible decision — it erases information about the path not taken. The Landauer cost of each F application is the entropy exported to the environment. The dm³ convergence rate μ = −2 is the rate of information processing per unit time: 2 bits per Lyapunov time.