||A philosophy of science originally proposed by the French philosopher and sociologist Auguste Comte (1798-1857) whose primary purpose was to distinguish science from both metaphysics and religion. Comte drew upon the earlier ideas of the French social reformer the Comte de Saint-Simon (1760-1825), and presented his own ideas in a series of lectures which were codified in The course of positivist philosophy (6 vols, 1830-42). The subsequent history of positivism in both philosophy and the sciences is complex, and there are several different versions of positivism (Bryant, 1985; Kolakowski, 1972). In the most general terms, however, Comtean positivism determined the scientific status of its statements through five steps:
(1) Scientific statements were to be grounded in a direct, immediate and empirically accessible experience of the world, and observation statements were therefore privileged over theoretical ones (see empiricism): observations of events were the leading particulars of scientific inquiry and as such, observation statements could be made independently of any theoretical statements that might subsequently be constructed around them.(2) Scientific observations had to be repeatable, and their generality was to be ensured by a unitary scientific method that was accepted and routinely drawn upon by the scientific community as a whole.(3) Science would advance through the formal construction of theories which, if empirically verified, would assume the status of scientific laws.(4) Those scientific laws would have a strictly technical function, in that they would reveal the effectivity or even the necessity but emphatically not the desirability of specific conjunctions of events: in other words, they had to take the form \'If A, then B\' (laws of constant conjunction or \'Humean laws\': for example, \'If somebody receives favourable information about an innovation, then they will decide to adopt it\'; see diffusion).(5) Scientific laws would be progressively unified and integrated into a single system of knowledge and truth (or, rather, Truth). The founders of the modern discipline of geography often relied on conceptions of science which were denied by positivism: Humboldt rejected the brute empiricism on which Comte\'s system was built, for example, and elaborated a sophisticated philosophical system in its place (Bowen, 1979). Yet much of the subsequent history of geography was dominated by the (often tacit) acceptance of some or all of these assumptions, so that when they were formalized during the quantitative revolution of the 1950s and 1960s the so-called \'New Geography\' that resulted was \'less of a radical departure than a logical extension of ideas which were already generally accepted by many geographers\' (Guelke, 1978; see also Gregory, 1978). Insofar as these were in any sense philosophically novel, they derived their originality from a commitment to logical positivism. The introduction of the prefix \'logical\' signalled a break with the classical Comtean model which had been made by the Vienna Circle, a group of philosophers, mathematicians and natural scientists that met regularly in Vienna in the 1920s and 1930s and whose manifesto â€” The scientific conception of the world: the Vienna Circle â€” was published in 1929. The basis of logical positivism was a distinction between:
(a)Â analytic statements: a priori propositions whose truth was guaranteed by their internal definitions and connections. These constituted the domain of the formal sciences, logic and mathematics, which (strictly speaking) had no place in a Comtean model that required all scientific statements to rely on empirical verification; and(b)Â empirical or synthetic statements: statements about things of all kinds whose truth had to be established empirically through hypothesis testing. This \'principle of verification\' was supposed to be the hallmark of the factual sciences, but it was soon challenged by Popper\'s principle of falsification. Instead of making every effort to prove an hypothesis, Popper believed that the scientist\'s task was to try to disprove it: if it stood up, then the hypothesis could be accepted, but only for the time being (see critical rationalism).These revisions provided empirical inquiry with a more secure basis than the classical Comtean model, and the \'New Geography\' was readily accommodated within their framework: the philosophical highpoint of this tradition was David Harvey\'s Explanation in geography (1969). The major omission within this conception of geography was the development of deductive-nomological explanation and a considered recognition of the importance of Popper\'s proposals for the business of hypothesis testing: certainly, much of the early work in spatial science depended on inductive-statistical rather than deductive-mathematical methods (Wilson, 1972). Even so, the overriding concern with the derivation, validation and integration of general theories of spatial organization was an unequivocally positivist project (see location theory). The debate over exceptionalism had established the legitimacy of a nomothetic conception of geography, and the search for models of spatial structure and methods of spatial analysis was soon widened by the incorporation of models and techniques drawn from other sciences that were also predicated on an unflinching positivism, e.g. classical mechanics and neo-classical economics. When research effort in human geography moved on from spatial form towards the study of generative process, much of its energy was still contained within the positivist tradition: e.g. the development of behavioural geography. But the discipline eventually broke through those bounds to confront a series of rival philosophical systems.
The critique of positivism that followed was both fierce and decisive: \'almost every aspect of the prescribed method can be shown to be open to question\' (Bowen, 1979; cf. Hay, 1979). And yet it needs to be emphasized that the relation between positivism as a philosophy and the evolving empirical method of human geography has remained contentious. Harvey\'s (1969) philosophical codification of spatial science and spatial analysis was largely retrospective: much of the innovative groundwork had already been done by geographers (including Harvey himself) who at the time had little interest in or even knowledge of philosophy, and Explanation in geography consistently referred to \'the\' scientific method rather than to positivism. In any case, empirical methods are constituted by more than philosophical warrants, and a revisionist historiography of geography informed by \'science studies\' is beginning to uncover the complex cultural history in which the quantitative revolution was embedded (Barnes, 1998; Hepple, 1998). Whether \'quantitative\' is an effective summary description of the \'revolution\' of the 1950s and 1960s is itself open to question, but there are also serious issues surrounding the connection between positivism and quantitative methods: positivists are not required to make use of formal mathematical or statistical methods, and neither does the use of quantitative techniques immediately identify (or condemn) an inquiry as incorrigibly positivist. None of this protects spatial science from the critique of positivism, but it does help to explain why so many quantitatively minded geographers were left unmoved by the philosophical interrogation of their work (Flowerdew, 1998). If, as Flowerdew says, the practice of quantitative geography continues to thrive, there is nevertheless no doubt that the critique of positivism had and continues to have a series of major effects on the contemporary discipline.
That critique has involved an assault on four plans of the positivist platform: empiricism, exclusivity, autonomy, and universality.
Empiricism: The relationship between observation statements and theoretical statements has been shown to be much more problematic than positivism allowed. The belief that \'the facts\' could speak for themselves was called into question in several ways. Within quantitative geography alternative modes of statistical inference were deployed in order to allow for varying degrees of theoretical co-determination, e.g. Bayesian theory. A recognition of the constitutive importance of \'theory\' reached far beyond formal analytical methods, however, and human geographers developed a much greater sensitivity to the importance of theoretical critique â€” to the clarification of concepts and categories â€” which prompted them to engage with theoretical systems that promised to provide a more incisive exposure of the structures of social life in time and space. In the first instance these theoretical systems were underwritten by foundational epistemologies like realism, but in the second instance non-foundational epistemologies like post-structuralism have risen to prominence. Post-structuralism has redoubled the significance of a critical interrogation of concepts and categories, but it is no theoreticism: it requires a close and careful reading of empirical materials to disclose their silences and erasures and their assumptions and ambiguities (cf. deconstruction).
Exclusivity: The assumption that the \'objective\' methods of the natural sciences could be extended into the domain of the humanities and the social sciences to provide a self-sufficient and unitary system of inquiry has been challenged. On the one side, the application of qualitative methods showed the importance of recovering meanings, intentions and values as an essential moment in human geographical inquiry: \'data\' could thus not reasonably be limited to the computational and enumerative (see humanistic geography; phenomenology). It would in any case be a mistake to think in terms of a simple binary opposition between the quantitative and the qualitative. It is possible to work with qualitative materials in thoroughly positivist ways and, equally, the rigorous use of quantitative methods can (and usually does) require a careful and critical understanding of the qualitative matrix in which they are embedded. On the other side, therefore, such supposedly \'subjective\' concerns were found to be important in the constitution and conduct of the natural sciences too, whose inquiries also involve difficult questions of meaning, interpretation and rhetoric (see naturalism; science, geography and). The privileges accorded to \'the\' scientific method were thus comprehensively compromised, and the natural sciences â€” whose power and importance was not in question â€” came to be seen in new, more complex and more interesting ways that nonetheless removed them from the pedestal on which positivism had placed them.
Autonomy: It was no longer possible to claim that science was (or should be) insulated from social life, offering \'neutral\' and \'value-free\' knowledge, and the embeddedness of scientific inquiry in social life was seen to require an explicit reflection on the ethics and politics of human geography: on its various \'regimes of truth\' and the ways in which they are imbricated in wider relations of power and knowledge (see critical theory; discourse; epistemology; ideology).
Universality: Even the natural sciences could be shown to be context-dependent in various ways, so that the extension and generalization of their findings becomes a precarious, negotiated achievement and not a given (Rouse, 1987). The same applied a fortiori to the humanities and social sciences, and in consequence human geography has been obliged to recognize that explanations and understandings arrived at in one setting cannot be transferred to other settings without critical scrutiny and often substantial reconstruction (cf. travelling theory). Human geographies are always \'local knowledges\' (cf. Barnes, 1989), even when they address global issues, and the recognition of all geographies as forms of situated knowledge means that it is vitally important to attend to the play of different interests from different positions and in different voices (see postmodernism).
What these philosophical counter-claims amounted to was a reinstatement of the social foundations and responsibilities of intellectual inquiry and a refusal to separate \'science\' from discourse more generally. Many human geographers no doubt continue to think of themselves as social scientists; many do not (the connections between human geography and the humanities are now probably stronger than at any time since the Second World War). In either case, however, probably very few count themselves as positivists. The continued exploration of post-positivist geographies has: (a) dimmed the old enthusiasm for the unrestrained application of quantitative techniques as ends in themselves (though analytical methods continue to be important in many areas of non-positivist geography); (b) reinvigorated a number of areas which had hitherto been characterized by an unreflective positivism; and (c) helped to open up new avenues of inquiry throughout the discipline and beyond (e.g. feminist geographies; post-colonialism). In retrospect, it seems clear that, in the most general terms, the critique of positivism at once shook the sense of certainty that the conjunction of positivism and spatial science had promised human geography â€” replacing it with a pervasive \'cartographic anxiety\' (Gregory, 1994) â€” and, since so much of physical geography remained largely unaffacted by and indifferent to these arguments, also made the connections between the disciplines of human geography and physical geography much more tenuous.Â (DG)
References Barnes, T. 1989: Place, space and theories of economic value: contextualism and essentialism in economic geography. Transactions, Institute of British Geographers NS 14: 299-316.Â Barnes, T. 1998: A history of regression: actors, networks, machines and numbers. Environment and Planning A 30: 203-23.Â Bowen, M.J. 1979: Scientific method â€” after positivism. Australian Geographical Studies 17: 210-16.Â Bryant, C. 1985: Positivism in social theory and research. London: Macmillan.Â Flowerdew, R. 1998: Reacting to Ground Truth. Environment and Planning A 30: 289-301.Â Gregory, D. 1978: Ideology, science and human geography. London: Hutchinson.Â Guelke, L. 1978: Geography and logical positivism. In D.T. Herbert and R.J. Johnston, eds, Geography and the urban environment, vol. 1. New York: John Wiley, 35-61.Â Harvey, D. 1969: Explanation in geography. London: Edward Arnold.Â Hay, A.M. 1979: Positivism in human geography: response to critics. In D.T. Herbert and R.J. Johnston, eds, Geography and the urban environment, vol. 2. New York: John Wiley, 1-26.Â Hepple, L. 1998: Context, social construction and statistics: regression, social science and human geography. Environment and Planning A 30: 225-34.Â Kolakowski, L. 1972: Positivist philosophy: from Hume to the Vienna Circle. Harmondsworth: Penguin.Â Rouse, J. 1987: Knowledge and power: toward a political philosophy of science. Ithaca: Cornell University Press.Â Wilson, A.G. 1972: Theoretical geography: some speculations. Transactions, Institute of British Geographers 57: 31-44.