No abstract

Contingency, Realism, and the Expansion MethodFor most disciplinary historians, the growth of spatial science in rography had its origins in a number of decisive critiques of the purposes an methods of the areal differentiation tradition that had previously dominated the discipline. Of these critiques, the one Schaefer (1953) directed against "exceptionalism" perhaps had the greatest impact. To Schaefer, the idea that geography was an exceptional discipline-one that remained methodologically unique from other social and natural sciences-led practitioners of regional geography to employ an idiographic approach that focused on the unique characteristics of areas. Schaefer held that this approach was bankrupt inasmuch as it had failed to lead to systematic generalizations from the mass of information regional geographers collected about particular places.By contrast, the proponents of the new spatial science argued in favor of a nomothetic ap roach that would explain the diversity of places through the 1953). Such laws were to be discovered through theory construction, model building, and mathematical and/or statistical analysis. Geographical Analysis was founded on these arguments, signaled in no small way by its bold subtitle: "An International Journal of Theoretical Geography." From its inception, the journal promoted a nomothetic approach to geography.For those who adhere to the tenets of spatial science, there is probably little doubt as to which group prevailed in the debate over the relative merits of the general and particular positions (Burton 1963;Golledge et al. 1982;Gould 1979). In fact, even for geographers who acknowledged the usefulness of integrating the two (for example, Taaffe 1974), it was the general that assumed the dominant position in the binary hierarchy.

This paper describes an experimental study that attempts to uncover some of the reasons why community residents find mental health facilities undesirable as potential neighbors. Respondents were asked to evaluate a variety of public facilities by indicating how similar they are in terms of “noxiousness.'’They were also asked how close they would prefer to live to each of the different facility types. From the results of the study it is possible to suggest some alternative strategies for siting new mental health facilities, such as co‐locating them with other human‐service agencies or locating them within larger facilities.

This paper investigates the effect of proximity on attitudes towards mental illness. One outcome of the policies referred to as deinstitutionalization is an overall reduction in the social and geographical distance between members of the public and the mentally ill. For deinstitutionalization to be successful, it is important that residents in impacted communities become more accepting, both in a passive and in an active sense. A survey was made of residents in two neighborhoods of the city of Norman, Oklahoma, one of which was adjacent to a large mental health facility. In a causal model framework, the relationship between proximity and public attitudes is investigated; and the results suggest an important and significant relationship. The implications of the study for public education campaigns about mental illness, and for facility location strategies, are discussed.

This paper uses a geographic shift-share model to examine the uneven development of manufacturing at the state and local scales in nine southeastern states from 1950 to 1990. The research compares changes in employment between states and among six types of local areas within each state. The types of local areas studied are: central city areas, urban fringe areas, large towns, small towns, rural nonfarm areas, and rural farm areas. The results indicate that: (1) the development of manufacturing in the Southeast was geographically uneven from 1950 to 1990 at both state and local scales, but more so at the local level; (2) development at the local scale shifted from rural nonfarm in the 1950s, to small and large towns in the 1960s, to urban central and fringe areas in the 1970s, and to a wide range of local areas in the 1980s; (3) development became less uneven between states over time for all types of local areas, and by the 1980s the degree of uneven development closely corresponded to the degree of urbanization (ruralization) of the local area, with small towns and rural areas being the most evenly developed and urban central areas being the most unevenly developed; (4) development became less uneven among the six types of local areas within every state over time, with Alabama, Kentucky, North Carolina, and Virginia being the most evenly developed and Florida, Mississippi, and South Carolina being the least evenly developed by the 1980s; and (5) labor costs were a significant determinant of the uneven development of manufacturing at the local scale, but the dominance of this factor appears to have subsided in the 1980s.