The Solar Pendulum Revisited: Changes in Solar Activity and the Goals of Successive Governments, 1930–2024

If solar cycle 25 marks the start of a new rising phase of the roughly century-long Gleissberg envelope, does the past tell us what kind of political leadership to expect over cycles 25 to 28? We tested three versions of the Solar Pendulum hypothesis on heads of government who took office between 1930 and 2024. The level version holds that high solar activity favours agentic leaders and low activity communal ones. The envelope version holds that strong eras of solar activity favour agentic leadership for decades. The homeostatic version holds that each change in solar activity between one government and the next nudges the successor in the direction of that change, so that societies correct course cycle by cycle. Leaders' goals were rated blind by two large language models, ChatGPT and Gemini, using a fixed agency–communion definition, and matched to the 13-month smoothed sunspot number of the World Data Center SILSO. Each test was written into a dated analysis plan before it was run, and the G20 tests before any G20 rating existed. Across three samples (62 pilot leaders selected with AI assistance, 195 leaders from ten countries chosen by rule, and 262 leaders from the G20 countries), leadership style showed no solar link, and the strength of the solar cycle showed none either (Spearman ρ = −0.02 in the G20 set). The homeostatic correction appeared in all three samples: in the G20 set, successors moved toward agentic goals in 47% of successions when solar activity had risen and 33% when it had fallen (ρ = 0.17, one-sided within-country permutation p = .013). The alignment is specific to the real solar timeline: when the sunspot series was shifted in time by 2 to 181 years, none of 2,144 misaligned versions of solar history fitted the G20 successions as well (about 80 effectively independent comparisons, p ≈ .01). The effect is modest but consistent, and it did not weaken during the last rising phase of the envelope. These results refine the original Solar Pendulum, moving the effect from leadership style to goals and from levels of solar activity to changes in it, as that study's remark on the cycle's directional momentum anticipated. Our projection is therefore a pendulum rather than a plateau: successors will lean communal as cycle 25 declines toward about 2030 and agentic as cycle 26 rises. Keywords: solar activity; Gleissberg cycle; political leadership; agency and communion; homeostasis; large language models