Previous Article |
Table of Contents
| Next Article 
Spatial Demography Special Feature
INTRODUCTION
Spatial demography
Department of Demography, University of California, Berkeley, CA 94720
This issue of PNAS brings together a collection of seven papers representing work in emerging areas of spatial demography, with special reference to North America. The papers have been assembled through the efforts of Susan Hanson of Clark University (Worcester, MA) and W. A. V. Clark of the University of California, Los Angeles.
Although place, density, and movement have always had a part in population studies, they have not often held center stage. Demographers' prime coordinates are the two time-like coordinates of time and age. The three space-like coordinates of physical location and the many-dimensional coordinates of social location have tended to play supporting roles. Today, this situation is changing, and a host of scientific questions that intermingle geography with demography and with the whole range of the social sciences are coming to the fore.
Two reasons for the expansion of spatial demography are readily apparent. One is the recent availability of finegrained spatial data linking geographic coordinates and categories to demographic, social, and economic variables, suited for analysis with the new computing tools of geographical information systems. In the United States, the Census Bureau's TIGER (Topologically Integrated Geographic Encoding and Referencing) system of the 1990s opened up opportunities that the private sector quickly augmented, and parallel developments have proceeded around the world. A second reason, in the United States, is "adversarial legalism" (1). Political decisions regarding distributional equity across jurisdictions, involving representation, public housing, discrimination, and civil rights, are propelled into the courts, creating and funding a demand for expertise in spatial analysis.
A deeper reason, perhaps, is an emotional recognition that a once-rich diversity of local particularities in customs, accents, values, legends, architecture, instincts, foods, and memories is vanishing, or retreating into less visible forms. Our technological ability to analyze differences from town to town, valley to valley, state to state, and freeway to freeway becomes greater at a time when sameness is replacing difference. We begin to value what we begin to lose.
"Migration" is the usual label for the spatial subfield of demography, and movement is its preoccupation. But values of place are often associated not with movement but with roots. Rootedness to place, webs of local social ties, place-specific knowledge and assets, and "knowing and being known" are themes in the first paper in this collection, by Susan Hanson (2). Hanson studies locational stability, and she highlights it with an illustrative analysis of its role in the founding of small businesses. Most new businesses are homegrown, making it especially interesting to study the exceptions: entrepreneurs who move and set up businesses where they have not lived before. The suspense in the study is to learn to what degree these individuals are truly newcomers or to what extent some set of less apparent ties is still providing local social capital.
Hanson compares interview data from Worcester, MA, and Colorado Springs, CO, a well matched comparison along many relevant dimensions. But it is poignant to think of such a comparison a hundred years ago, when cultural and social divides would have set the two places utterly apart. Economic strategy and rational choice ought to be in view when the informants are businessmen and businesswomen, but culture, although out of view, may also have some say.
Hanson's paper deals with measurement and with consequences of stability, not with culture and rationality. Indirectly, however, it reminds us how outlooks can differ. Coming from contrasting academic traditions, French spatial theorists and American authors have different takes on the salient constituents of locational stability. The great French demographer Hervé LeBras, in books extending from L'Invention de la France to La Planète au Village (3, 4), has applied tools as far-ranging as fractal geometries for understanding persisting political, economic, and behavioral gradients as geographical cultural phenomena.
The second paper in the collection, by W. A. V. Clark (5), turns to the smaller unit of neighborhoods, to urban concentrations of poverty, to government housing programs, and to the choices of people who "vote with their feet." Clark analyzes data released so far from the Moving to Opportunity Program of the U.S. Department of Housing and Urban Development, which includes experimental randomization of families granted different forms of vouchers for housing assistance. Clark focuses on Baltimore and considers both the initial move of program participants, affected directly by program incentives, and subsequent moves. Some see the results of the Moving to Opportunity study as the way forward; Clark, however, is more cautious and suggests that these interventions may not necessarily yield outcomes that will change residential patterns. The existence of experimental data, despite inevitable limitations, presents a great opportunity for spatial demographers to contribute scientific findings to a national policy debate.
Plane et al. (6), in the third paper, hearken back to the origins of systematic migration studies, examining migration flows up and down an "urban hierarchy" of town and city size. The late Victorian "Laws of Migration" enunciated by E. G. Ravenstein portrayed movements in short successive steps from villages to towns, towns to cities, and cities to the capital, reflecting processes of progressive industrialization. Today, in contrast, in the United States, the authors show that steps down the urban hierarchy predominate and that the largest flow is a leap down the hierarchy from "megacities" to the nearest thing to rural areas in the classification, which bear the awkward name of "non-core-based statistical areas."
For the tabulations of Plane et al. (6), all megacities are put together in one category, all "micropolitan areas" in another, and so forth. But particular members of each category are, of course, linked together in core-periphery systems, whose far-ranging impacts on social behavior have been studied by G. W. Skinner (7) in the context of China, Japan, and Europe, providing a framework for further study here.
Urban hierarchies are a tricky topic, because shifting population keeps producing shifting classification. The distributional changes measured by the researchers are being played out in the midst of change in the aggregate, a doubling of the U.S. population since 1950. In the face of such growth, how should we translate positions in the urban hierarchy into corresponding personal experience? Are we to think of the same kinds of places, all just proportionately larger? Or should we think of the kinds of places found within the urban hierarchy of the last generation disappearing and being replaced by places of new kinds?
Migration across levels in the urban hierarchy, as Plane et al. (6) point out, are highly specific to stages in the life course. Rogerson and Kim (8), in the fourth paper, study such transitions in detail. They trace the movements of the baby-boom cohorts through space, age, and time. Transitions through age are bound up with family relationships, now with care and support for elderly parents and soon with baby boomers' own needs for family care and support. The literature on kinship resources based on demographic computer microsimulation (9), which emphasizes step-children as well as biological children, stands to profit from the opportunities for geographical tracking demonstrated in this paper.
The United States has the largest immigration rates of any large country in the world. Changes in spatial distribution and social arrangements necessarily direct attention toward immigrant experience. Most studies tabulate immigrants as individuals. The fifth paper, by Ellis and Wright (10), takes a different tack, examining households by immigrant composition or, more cogently, by the mixture of new immigrants, second-generation immigrants, and residents with more generations of residence in this country behind them. A notable finding is the substantial number of households that combine immigrants with third-generation members, putting assimilation processes in a different light.
The most visible arena in which spatial demography intersects with public policy is the process of redistricting, in tugs-of-war between political parties as well as between state legislatures and the courts. In the sixth paper, Benjamin Forest (11) considers the technological advances with computer-based boundary construction. Scientific capacities would seem to offer an alternative to recurrent political conflict, enabling optimization in accord with specified objective criteria. But Forest shows how the criteria established by the courts are inherently contradictory, so that, perhaps by design, any scientific approach is kept vulnerable to litigation.
The concluding paper in the collection, by VanWey et al. (12), reviews problems of data confidentiality. Spatial coordinates augment the risk that survey or study respondents might become identifiable, despite careful application of established methods to delete personal identifiers and protect anonymity. New solutions will have to be found for these problems if the potentials for linked spatial and social data are to be realized.
The seven papers in this collection provide a taste of current work in spatial demography, but the field is necessarily broader than the topics covered here. Demographic methods deserve mention. The groundwork for modeling and projecting spatial transitions was laid by Andrei Rogers (13), and stochastic geometry was pioneered by David Kendall (14). The versatility of spatial modeling is well illustrated by LeBras (4). Recent contributions in PNAS include a remarkable algorithm for constructing thematic cartograms based on diffusion equations from physics applied to U.S. voting patterns (15) and a paper marshalling approaches from network analysis to describe the connectivity of the world airline transportation network (16).
The natural disasters of the present year have drawn attention to the value of geographical information systems for forecasting and identifying populations at risk. It proved possible to compute estimates of total population in lowlying areas vulnerable to the tsunami of Dec. 26, 2004, almost as quickly as the waves were ebbing away around the globe (17).
Turning our attention from hours to millennia, we find spatial demography contributing to our understanding of the long-term evolution, dispersal, and kinship of human populations. The relative roles of technological diffusion and of demic, migration-based diffusion in the spread of agriculture continue to attract productive debate informing popular scientific writing (ref. 18, pp. 146-154). Advances are being made in models from probability theory that tie spatial structure to the reconstruction of genealogies and phylogenies from DNA (19). Small migratory flows over long time periods can have large implications. Drawing on anthropological opinion about flows between particular nearly isolated populations across prehistory and history, Rohde et al. (20) have estimated that the most recent ancestor that every living human has in common, through male or female lines, could have lived as recently as A.D. 55.
Spatial demography extends across space and time, from the changing face of the everyday world of present experience to glimmerings of our remote origins and interconnections. It brings sciences together: geography and demography, political and social sciences, mathematics, statistics, physics, and biology. In California, Napa Valley vineyards have started to offer tastings not as a "sampler" but as a "flight" of wines. In this spirit, PNAS offers a flight of seven tastings from spatial demography.
 |
Footnotes |
|---|
* E-mail: wachter{at}demog.berkeley.edu.
© 2005 by The National Academy of Sciences of the USA
|
References |
|---|
 Top References |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
