Positivism A Poor Model for Scientific Geography

As the previous paragraph implies, the answer depends entirely on whether one operates with a normative or an empirical definition of science. The former defines science in terms of an ideal-type terms and then judges the actual conduct of scientists against the ideals. The latter is more pragmatic and can be captured thus: 'science is what people who call themselves scientists actually say and do'. Sometimes, the first definition of science is confused with the second. This was the case with the first full-blooded criticisms of spatial science in the 1970s, notably those of Gregory (1978) and Guelke (1971; 1979). Together, these authors created the impression that spatial science was explicitly positivist in both world-view and method. It is not possible to rehearse all of the above author's criticisms. I will just offer a sampling of their complaints in relation to the world-view/method/quantification trinity that supposedly underpinned that search.

First, to take one of Comte's five principles of science, Guelke and Gregory pointed out that the spatial scientists often failed to respect the first, i.e. that scientific knowledge was based on direct experience and observation of reality. For example, Bunge's Theoretical Geography (1962) deployed geometry as a descriptive language that might depict - albeit approximately - real spatial patterns. Yet his clean geometric lines bore little relation to the complexities of real physical and human landscapes. The elegance of geometry seemed to take precedence over what the evidence suggested was actually happening. Second, Gregory and Guelke identified problems with geographers' use of the deductive-nomological procedure. For instance, Guelke argued that no laws could be found to describe human behaviour because people do not behave in regular, law-like ways. Yet human geographers during the 1960s frequently tried to explain human geographic behaviour as examples of (or 'deviations' from) a core rationality that all people were said to possess. Gregory, meanwhile, argued that many spatial scientists' laws, theories and models were descriptive rather than explanatory. They often took the form of identifying regular associations (or correlations) among certain phenomena (e.g. high rainfall events and floods), yet without offering real explanations of whether these associations were accidental or causally connected. To add to all this, Guelke pointed out that geographers cannot usually exert experimental control over the things they wish to study. Where the laboratory sciences can isolate the variables whose causal relations are of interest, geographers must study large-scale, complex and often changeable 'open systems'. This, Guelke, maintained, can make hypothesis testing very difficult: here 'the goal of investigation becomes . . . expla[ining] . . . the discrepancies between theoretical constructs and reality, rather than the explanation of reality itself' (Guelke, 1971: 49). Finally, Gregory and Guelke both suggested that spatial science's obsession with measuring spatial patterns had produced a lot of precise factual information as if this was an end in itself. The impulse to measure became disconnected from the wider goals of positivist world-view and method in their view. As Holt-Jensen (1995: 829) observed in a retrospective essay, 'spatial science research developed greater refinement of description rather than explanation'. Guelke thus concluded that spatial science was saddled with the double embarrassment of having sets of untestable/simplistic laws, theories and models, on the one hand, and masses of confusing empirical data, on the other.

These were devastating criticisms but many geographers persisted with some or all of the elements of the positivist approach to geographical inquiry. This was not unreasonable. A central problem with criticisms such as those of Guelke and Gregory is that they measured spatial science against an ideal-type that supposedly dictated what 'real science' looked like. But this normative critique was and is problematic because it is unclear why philosophers like Comte and the Vienna Circle were entitled to dictate the nature of science to practising scientists. In any case, the Vienna Circle drew some of its inspiration from how physicists did their research, raising the question of why a mainly laboratory science should form the benchmark of scientificity for a mainly field subject like geography.

In light of this, an empirical critique of spatial science suggests itself. Such a critique looks at what spatial scientists actually did and evaluates their activities within the specific disciplinary context that was post-war geography. In this approach, there is no question that these researchers were not fully or properly scientists. Because they called themselves 'scientists' we look at the actual research conducted under this banner and evaluate it in its own right and also in relation to alternative approaches in geography advocated at that time. Again, it is not possible to do this in detail, but two illustrative points can be made. First, though Guelke and Gregory were right that spatial scientists often produced few compelling explanations of the phenomena they measured, description per se is often illuminating and useful. Many of the spatial associations discovered in the 1960s and 1970s were hitherto unknown and could be acted upon. For example, knowing the exact number of vehicles in rush-hour traffic in a given city day-in, day-out, is central to knowing whether it is necessary to build relief roads or to introduce extra park-and-ride schemes. Explaining the traffic volume-time-of-day link is, strictly, immaterial (though, of course, interesting and potentially useful). Second, in relation to the regional geography that carried on despite the Schaefer-Hartshorne debate, spatial science research certainly produced more general and precise understandings of the geographical landscape. Descriptive or not, these understandings allowed geography to gain wider social respectability and gave geographers a role in environmental management and urban and regional policy.

Was this article helpful?

0 0


  • Amanuel Hagos
    What are poor models science?
    8 years ago

Post a comment