This is amusing and apt:

[P]erhaps the civil servants, accountants and managers could benefit from better interpretations of their activities and words. An ironic example is the demand in the UK that university departments demonstrate the “impact” of particular pieces of research over a relatively short timescale and excluding effects on students, or even sales of books. “Impact” is a term drawn from mechanics, where it implies a particular kind of causation, a definite event giving some other identifiable thing a shove or a biff. Its magnitude can be quantified and, of course, if quantitative methods are all you are allowed to use, there is a temptation to suppose that everything else can be quantified as well: to him whose only tool is a hammer, everything looks like a nail. But improved understandings do not work by shoving or biffing. They can seldom be traced back to one particular essay or one moment in time. Ideas work by osmosis, filtering into people’s minds over long periods, manifested in innumerable subtle changes of thought and behaviour seldom attributable to just one antecedent event. A better model would be Don Basilio’s excellent understanding of the cumulative effect of gossip in his great encomium to the power of calumny in Rossini’s The Barber of Seville. Perhaps we might hope that even if they do not care to wrap their minds around issues in the philosophy of causation, our masters might be encouraged to attend an evening of opera buffa.