Introduction

People love new tools that enable them to do what they have dreamed of doing. Passionate engagement with methodology is the academic expression of this basic human characteristic. And the buzz we feel when we are able to share our enthusiasm for a new method is one of the joys of academic work. The 1998 and 1999 Social Science History Association (SSHA) sessions on historical GIS were exciting for just these reasons. In the best SSHA tradition, people in many different fields discovered their common interest in applying GIS (geographic information systems) to historical research. The sessions, which were organized principally by Humphrey R. Southall, generated the kind of excitement that typified the first SSHA meetings, when participants had “the sense of being on the cutting edge of scholarship” (Bogue 1987: 338)—of making something new by using new tools.

What is historical GIS? In structure, it is the same as other kinds of GIS, which all essentially consist of a spatial database that integrates map-based information about the historical location of certain entities (such as census districts, industrial firms, or rivers) with quantitative or qualitative information about those entities (such as population, product, or level of pollution).¹ The key difference between historical GIS and the vast majority of GIS practiced today is that its source data typically include archival material that must be converted from analog to digital form. The scanning and digitization of historical material, including maps (Cobb 1999),² is already blurring this distinction, and the accumulation of digital spatial databases is creating opportunities for GIS analyses of the recent past, particularly in urban, transportation, business, and environmental history. But the problem of converting the bulk of paper historical sources into digital form, particularly unsystematic qualitative sources, poses special challenges for historical scholars, as does the problem of locating historical places and objects in order to assign them the geographical coordinates required by a GIS.

The more important question to readers of this journal is, what good is historical GIS as a method for social science history? The answer, illustrated by each essay in this volume, is that historical GIS makes space an explicit part of analysis. It extends quantification and systematic empirical analysis to questions, scales, and evidence that few historians have considered. The overarching argument of this collection of essays is that space should be an explicit part of historical analysis because:

1. Accurate spatial boundaries are key to all calculations derived from geographically located data. Without using historically correct and accurate unit area boundaries, one cannot tell whether statistical changes reflect changes in population, changes in boundary lines, or both.
2. We poorly understand the spatial aspects of human history, from the changing importance of physical geography as a constraining and enabling factor to the geographical expression of social structures.
3. Mapping data reveals dimensions of historical reality and change that no other mode of analysis can reveal.

Space and its human manipulation in place are the central subjects of historical geography (see, e.g., Conzen 1990; Earle 1992). Historical GIS computerizes much that is basic to historical geography by making spatial objects—that is, anything that exists in geographical space—and their spatial qualities and relationships the basis for computer analysis.

The contributors to this special issue of Social Science History are part of a groundswell of interest in historical GIS. As often happens during the early phase of a methodology’s employment, most were unaware of one another’s work until they met at the 1998 SSHA meeting in Chicago. The interest generated at that meeting resulted in another, larger round of sessions on historical GIS at the 1999 conference. Half of the participants in those sessions had never before attended an SSHA meeting, which added to the sense of mutual discovery. While the essays in this volume present some of the best work from the 1998 sessions, they also represent many of the practical, theoretical, and substantive issues that arose in spirited discussions the following year. The contributors and I hope...