The place of locational analysis: a selective and interpretive history

The paper has two purposes. The first is to provide a selective review of the history of locational analysis as it bears on economic geography. Three periods are examined: the German location school of von Thunen, Weber and Losch that begins in the first part of the nineteenth century and ends in the middle of the twentieth century; American spatial science that starts in the mid-1950s and is in decline by the late 1970s; and the new economic geography associated with the economist Paul Krugman, and inaugurated by his 1991 book, Geography and trade. The second is to make a methodological argument. Locational analysis is most frequently justified in terms of the purity of its logical and mathematical reasoning, permitting some commentators to trace an unbroken line of progress over its 175-year history. I argue that such a claim is based upon acceptance of a broader philosophical position, rationalism - the belief that the foundation of knowledge is reason - which takes its most perfect form in logic and mathematics. Rationalism has been criticized in various ways ever since it first emerged in the Enlightenment, however, and the same critical sensibility informs this paper. Specifically, I criticize the rationalist interpretation of locational analysis by drawing upon a recent interdisciplinary body of literature arguing for the importance of local knowledge, the idea that knowledge, even abstract theoretical knowledge of the kind found in locational analysis, is shaped not by the universal but by the peculiar historical and geographical context of its production. This anti-rationalist argument in favour of local knowledge is exemplified by discussion of the three periods of locational analysis that form the paper's core.