When the builder of the world's first steamship, Henry Bell, sent the British Admiralty details about his ideas for this epoch-making maritime development, their naval Lordships treated the information with ‘cold neglect’, seeing steam propulsion as irrelevant to the future of naval warfare (Mitchell 1971: 131). When information emerged about a near-meltdown at the US Three Mile Island nuclear power plant in 1978, engineers and managers at the Soviet Chernobyl nuclear plant dismissed it as a product of capitalism's inherent tendency to sacrifice safety for profit (as compared to in the Soviet Union, where no nuclear accidents were ever reported), and declared that their own installation could have nothing to learn from the US failure (Hawkes et al. 1986: 110). When, only a few years after the collapse of the Soviet Union, Tony Blair's New Labour Government in Britain adopted a system of targets to apply to every part of the public services, well-known problems of performance target systems – ratchet effects, threshold effects, output distortion – that had been carefully documented by Soviet historians and economists since the 1950s were wholly ignored, only to be rediscovered the hard way as all of those problems emerged in the British target system (Hood 2006).

So can knowledge and action be effectively brought together in government and public services? It's easy to slip into earnest clichés about the importance of evidence-based policy or better statistics for good governance and to gloss over the formidable social and cultural gulf that so often yawns between ‘knowers’ and ‘doers’.