As an imagologist I am often asked by my social-scientific or social-historical colleagues how I can determine the representativity of the literary material I study. If, in studying the English image of Italy, I draw on George Eliot and E.M. Foster, is that not a very restricted and rarefied data sample, almost a random stab in the dark? What wider conclusions could possibly be drawn from such a minute sample regarding ‘the’ image of Italy in England, or attitudes generally vis-à-vis Italy as current in England?
That ‘representativity’ challenge is irrelevant and pernicious, and should be rejected out of hand. It imposes on the humanities an entire set of assumptions and working methods that are alien and uncongenial to it, beginning with the idea (too ingrained to be consciously reflected upon, let alone queried) that what we study are selected samples as proxy data for a larger whole – that larger whole being, ideally, society as such.
[....]