Humanity's energy history can be read in three great transitions: from wood to coal (late 18th century), from coal to oil (mid-20th century), and now from fossil fuel to electricity — and from fossil-fired electricity to electricity from renewable sources. The first two transitions were driven by scarcity, war and politics; the third is driven by a cost curve. In 1976 a watt of solar capacity cost $79; in 2020 the same watt cost $0.20. Known as Swanson's Law, the curve has held: each doubling of cumulative production cuts module price by about 20% — a learning curve analogous to Moore's Law in semiconductors.

The critical landmarks span decades. The 1979 oil shock prompted the Carter administration to install 32 solar panels on the White House roof (Reagan removed them in 1986). Germany's 2000 Renewable Energy Sources Act (EEG) created scale through a feed-in tariff — demand subsidised by German consumers between 2004 and 2012 effectively built China's panel manufacturing sector. Through the 2010s China took over 75%+ of global solar panel manufacturing; when manufacturing scale met deliberate industrial policy, module prices fell off a cliff. Wind turbines grew in parallel: a typical onshore turbine in 2000 produced 1 MW, in 2024 it is 6 MW; offshore units have reached 15 MW (China's MingYang). A single 15 MW offshore turbine produces enough electricity for around 20,000 homes a year.

The 2020s became the turning point because cost, policy and storage came into line at once. Net-zero pledges — EU 2050, UK 2050, US 2050, China 2060 — together cover more than 90% of world GDP. The US Inflation Reduction Act of 2022 committed $369 billion to climate and clean energy investment — the largest single statutory clean-energy package anywhere. In 2023 China added 277 GW of solar and 76 GW of wind capacity (more than half of the world's total renewable additions). Renewables produced about 30% of global electricity in 2023 (14% hydro, 8% wind, 5% solar, 3% other); in 2024 solar for the first time added more annual capacity worldwide than coal. Battery storage costs fell about 90% between 2010 and 2024; lithium-iron-phosphate (LFP) chemistry made grid-scale storage economic.

The picture is nevertheless two-sided. Absolute fossil-fuel consumption was still at historical highs in 2024: oil ~103 million barrels/day, coal ~8.8 billion tonnes, gas ~4.1 trillion m³. Renewables are growing fast, but energy demand is also growing (AI data centres alone consumed about 1.5% of global electricity in 2024 and projections for 2030 range from 4-8%). Three bottlenecks remain. 1) Grid: building the high-voltage transmission that moves and balances variable sources turned out to take decades, not years (a typical US line permits 10-15 years). 2) Critical minerals: supply of lithium, cobalt, copper, nickel and rare-earth elements is failing to keep up with demand and is creating geopolitical dependencies (Congolese cobalt, Chinese rare-earth processing). 3) Storage: hours of storage are now economic, but weeks and months are not, so seasonal gaps still have to be plugged by gas, nuclear or technologies that have not yet matured (hydrogen, enhanced geothermal, fusion). To hold warming to 1.5°C the deployment rate needed in 2030 is about three times the present rate (IEA Net Zero Emissions scenario). The historical significance remains clear: in the early 2020s humanity began to leave behind the founding energy logic of the industrial revolution — more fuel equals more energy. From here on the questions are matters of speed, not direction.