Cause and Effect in Geographic Analysis: Anachronism or Useful Concept?

CAUSE AND EFFECT IN GEOGRAPHIC ANALYSIS: ANACHRONISM OR USEFUL CONCEPT? Orin C. Patton* The terminology of "cause and effect" has been regarded with great suspicion in geographic literature since the decline and fall of environmental determinism. (1 ) Hartshome tells us that in earlier days "Hettner, and other writers following him, including such phrases as 'causally related,' 'causal connections,' or 'differences . . . interrelated with each other,' " (2 ) The explanations of that era emphasized causes originating in the physical environment , and Emrys Jones has traced efforts to salvage some sort ofconnection between the environment and man's reaction to it through "possibilism," Griffith Taylor's "modified determinism," O.H.K. Spate's "probabilism," and A.F Martin's call for an "uncompromisingly scientific human geography based on cause and effect and rigorous scientific law." (3) In spite ofthe continuing calls for a more scientific geography emphasizing both non-human and human determinants, we rarely see the term in contemporary human geography , although it is common in scientific literature as well as in everyday speech. We apparently take a great deal of care to avoid the expression although there is an irreducible minimum of the basic notion whether we dress it in the language of "possibilism," "probabilism," or "empirical regularities." If we deny all possibility of anything akin to causal relations, ifthere are no observable regularities that justify the expectation that certain states, causes or forces are followed by certain other states or effects, then our analyses are meaningless. This paper contends that 1 ) the terminology of cause and effect, as well as the explicit use of the concept, can rest on respectable philosophical grounds, 2 ) this useage is methodologically viable, 3 ) this is a useful and natural approach to research with great heuristic values, and 4 ) it is compatible with recent restatements of the purpose and scope of geography. We will assume here that "scientific geography" (4) may be defined, after Yeates, as the "science concerned with the rational development, and testing, of theories that explain and predict the spatial distribution and location of various characteristics on the surface of the earth." (5) This definition, while not all-inclusive, was chosen to emphasize scientific methods with the purposes of constructing theories that explain and predict. The construction and testing of geographic theory will be the process ofidentifying and testing the *Major Patton is assistant professor of geography at the United States Air Force Academy, Colorado. The paper was accepted for publication in November 1969. The opinions expressed in this paper are those of the author and in no way reflect the views of the United States Government or the United States Air Force. Vol. X, No. 1 interrelationships among spatial variables, but our theories and causal interpretations apply explicitly to the models we use, rather than to the real world, to avoid the philosophical problems discussed below. PHILOSOPHICAL PROBLEMS OF CAUSE AND EFFECT. The questions of cause and effect relations have been among the more prominent themes ofwestern philosophy, but they have not yet been resolved. A list ofthe major contributors to the literature would read like a "Who's Who" in philosophy and science. Any concise summary of the literature is quite likely to do violence to philosophical subtleties, but an identification of the major issues is necessary. (6) The issues most directly relevant to modern philosophy were introduced in the attacks led by Hume and the empiricists on causality. (7) It was Hume's position that we are limited to sense experience and that we have no grounds for attributing cause and effect relations of any sort since we are limited to the experience of perceiving only "one object following another." (8 ) The "necessary connexion" linking the cause and the effect can never be experienced; the expectation of experiencing the effect when we experience the cause is a "species of natural instincts, which no reasoning or process of the thought and understanding is able either to produce or to prevent." (9) Thus, not even Hume denies the expectation ofthe effect, and he denies only the metaphysical link between the effect and its cause. The Humean "connexion " arises not from the observation of a single event "but when many uniform instances appear...