Edwards’s A Vast Machine offers a definitive account of climate science by rejecting the false opposition between models and data. His central proposition is stark: in global climate knowledge, there are no pure data, because observations become meaningful only through models, standards, instruments, archives, telecommunications and institutions that make them planetary in scope. The book’s title, drawn from Ruskin’s nineteenth-century image of meteorology as a “vast machine”, captures Edwards’s argument that climate knowledge is produced by a global knowledge infrastructure rather than by isolated observations. The visual materials sharpen this claim: the Apollo Earth photograph on page 31 crystallises the modern capacity to “think globally”, while the World Meteorological Organization diagram on page 34 shows the Global Observing System, Global Telecommunication System and Global Data Processing and Forecast System as interlocking technical and institutional circuits. Edwards’s case synthesis turns on the making of global climate records: weather stations, satellites, balloons, ships, models, metadata and reanalysis systems must be reconciled through laborious processes of data friction and infrastructural inversion. Climate data therefore “shimmer”, not because they are arbitrary, but because each reconstruction refines the past through evolving methods. Ultimately, Edwards demonstrates that global warming is known through historically sedimented infrastructures whose politics are everywhere: in standards, Cold War satellites, international organisations, model validation, IPCC assessment and the very possibility of imagining Earth as one climatic system.
