Quantitative Revolution

The quantitative revolution is the profound intellectual transformation occurring in Anglo-American geography beginning in the mid-1950s that followed from the use of scientific forms of theorizing and statistical techniques of description and empirical verification. In the process, an older regional geography concerned with describing, cataloguing, and delineating unique places was pushed aside and replaced by the “new geography” directed toward explaining, scientifically proving, and abstractly theorizing spatial phenomena and relations. Geography no longer was rote memorization of regional capitals, major waterways, and principal products; instead, it was now a science, that is, spatial science.

[Page 395]The quantitative revolution's origins are in World War II. Several of the quantitative revolution pioneers were first trained in statistical methods and scientific theory while serving in the military (particularly in the U.S. Air Force). In addition, wartime service convinced a number of geographers conscripted by the U.S. Office of Strategic Services (OSS, an important arm of military intelligence) of the limitations of the older regional geography. Necessary was geographic systematicity, explanatory purchase, and practical focus, none of which was found in the regional geography in which such geographers were trained. This experience proved to be decisive in postwar university classrooms.

It took roughly 10 years for the seeds planted during World War II to germinate. When the quantitative revolution emerged, it initially was highly localized and centered on one or two key individuals. In the United States, pivotal were the geography departments at the University of Washington in Seattle and the University of Iowa. At Washington, it was Edward Ullman (formerly at the OSS) and William Garrison who made the difference. In 1954, Garrison gave the first advanced course in statistical methodology in a U.S. geography department. And in an early advertisement for the department, the chair, Donald Hudson, boasted of the departmental use of an IBM digital computer, another national first. The first cohort of graduate students from that department (the “space cadets”) became a “who's who” of geography's quantitative revolution: Brian Berry, Ron Boyce, William Bunge, Michael Dacey, Arthur Getis, Richard Morrill, John Nystuen, and Waldo Tobler. Collectively, this group was critical in diffusing the Washington message and did so by rapidly establishing themselves and their research agenda at several prestigious U.S. universities, including the University of Chicago, Northwestern University, and the University of Michigan. At Iowa, Harold McCarty, the first human geographer to use a regression equation, was decisive. He attracted a number of graduate students who again were vital in spreading the word about numbers at places such as Ohio State University, MacMaster University, and (later) the University of California, Santa Barbara. Outside of North America, Peter Haggett and Richard Chorley in the United Kingdom (the “terrible twins” of British geography) and Torsten Hägerstrand in Sweden were crucial in establishing European beachheads.

The new geography that emerged, and that was solidified by the mid-1960s, was characterized by several features.

• A thirst for rigorous formal theory and slaked by begging, borrowing, and stealing from at least five sources outside of geography. First, Newtonian physics provided ideas of gravity and potential as well as the basis for spatial interaction modeling, that is, the analysis of geographic flows of people and things. Second, neoclassical economics gave the rationality postulate used to theorize geographic choice. Third, an older and hitherto forgotten German school of location theory offered mathematically exact models of agricultural land use, industrial location, and urban–economic settlement patterns. Fourth, urban sociology afforded both intra- and intermetropolitan explanatory models of population and their sociological characteristics. Fifth, geometry made available axioms of topology (the mathematical study of spatial forms) used in transportation studies. More generally, there was a belief that rigorous theory would reveal and explain an underlying spatial order and, at the limit, could be couched as a series of geographic laws of the type found in natural science.
• The use of an increasingly sophisticated set of statistical and mathematical methods. Initially, statistical hypothesis testing was rudimentary, but it was quickly ratcheted upward. By the mid-1960s, there was widespread use of complex multivariate inferential statistical techniques. In addition, there was pure mathematical modeling in which formal models were logically derived from a set of abstract assumptions expressed precisely. The implicit justification in both cases was that the geographic world and the world of mathematics were fundamentally ordered according to the same rational logic. Mathematics is nature's own language.
• A reliance on computerization. The first commercially sold computers were introduced on American campuses during the mid-1950s (the IBM 650 was the first and was introduced at Columbia University in 1954). Initially, there were no formal programming languages, and the capacity of computers to carry out calculations was limited. The pioneers of the quantitative revolution, however, were some of the earliest users of computers in American universities, often not getting their turn until after midnight and improvising programming techniques on the fly. By the mid-1960s, the computer was essential to the new geography. The complex calculations necessitated by multivariate statistical techniques and large-scale data sets could not be undertaken in any other way.
[Page 396]
• A new professional and social structure. Young, male, very ambitious, very able graduate students and junior faculty primarily forged the quantitative revolution. Initially blocked by a regional old guard, the “young Turks” set up their own dedicated conferences, their own training sessions to teach the rest of the profession the merits of a quantitative sensibility (the Summer Institutes in Quantitative Geography initiated in 1961 were formative), and their own outlets for publication (the various discussion paper series, particularly the Michigan Interuniversity Community of Mathematical Geographers [MICMOG], were crucial, culminating in 1969 with the specialized journal of quantitative geography, Geographical Analysis). In this sense, the quantitative revolution was as much a social and institutional transformation as an intellectual one.
• The emergence of an alternative philosophical justification for geographic research—positivism. For the most part, early proponents of the quantitative revolution did not understand their work in philosophical terms. By the end of the 1960s, however, considerable philosophical reflection had gone on around the quantitative revolution's larger intellectual justification. David Harvey's Explanation in Geography, published in 1969, was the culminating volume arguing that legitimating the quantitative revolution was logical positivism, a philosophy averring that true statements were those—and only those—in which logically consistent theory corresponded flawlessly to experientially grounded facts. It was a far cry from the vision of geography as rote memorization.

Only 4 years later, however, Harvey launched a counterrevolution based on Marxism that, within a decade or so, undid the quantifiers. Partly causing the quantitative revolution's fall was its inability to engage with pressing outside social and political issues and writ large during the late 1960s and 1970s around poverty, civil rights, the environment, war, and gender and racial equality. Partly also, there was a new generation of geographers entering the discipline who, like the quantifiers of the mid-1950s, wanted to make a distinctive mark. In their case, however, it was to be through social theory, not scientific theory. The continuity of that noun—theory—was significant. It ensured that human geography remained part of the social sciences, not lapsing back to the netherworld status in which it languished before the quantitative revolution. This is perhaps where the real revolution lay.

...