Introduction to Historical GIS and the Study of Urban History

Author(s): Donald A. DeBats and Ian N. Gregory

Source: Social Science History , Winter 2011, Vol. 35, No. 4, Special Issue: Historical GIS
and the Study of Urban History (Winter 2011), pp. 455-463
Published by: Cambridge University Press

Stable URL: http://www.jstor.com/stable/41407087

REFERENCES
Linked references are available on JSTOR for this article:
http://www.jstor.com/stable/41407087?seq=1&cid=pdf-
reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

Cambridge University Press is collaborating with JSTOR to digitize, preserve and extend access
to Social Science History

This content downloaded from

54.146.125.244 on Wed, 20 Nov 2024 15:17:53 UTC
All use subject to https://about.jstor.org/terms
Donald A. DeBats and Ian N. Gregory

Introduction to Historical GIS

and the Study of Urban History

Over the past decade or so geographic information systems (GIS) method-

ology has become an accepted tool in historical research (Gregory and Ell
2007; Knowles 2008). Although often regarded as a mapping tool, GIS is
perhaps better thought of as a type of database. What makes a GIS database
unique is that a location is stored for each item of data, with this location
taking any of a variety of forms: a point, a line, a polygon representing an
area or zone, or, in the case of a raster system, a pixel. GIS can then present
instantly on the screen a map showing the distribution of any variable or
combination of variables in any of the chosen locational formats. This elec-
tronic display of information becomes an analytic tool, allowing the refine-
ment of research questions, with answers displayed instantly: GIS creates
a display of information once visible only in paper form, drawn slowly and
expensively first by cartographers and then by vector plotters. GIS and its
associated tools transform mapping into a dynamic exploratory process.
The fact that the data in a GIS database are spatially referenced allows
a researcher to produce maps quickly, easily, and potentially in large vol-
umes; but a number of other advantages also make GIS a platform that is
well suited to the analysis of the geographies of the past. The first is that, as
all data in a GIS-informed project have an explicit spatial location, it is easy
to ask questions about where features are located in relation to each other. As
Anne Knowles (2000: 453) notes, the enhanced visualization component of
GIS has very positive results for historians, making available "dimensions of
historical reality and change that no other mode of analysis can reveal."

Social Science History 35:4 (Winter 2011)

DOI 10.1215/01455532-1381814
© 2011 by Social Science History Association

This content downloaded from

54.146.125.244 on Wed, 20 Nov 2024 15:17:53 UTC
All use subject to https://about.jstor.org/terms
456 Social Science History

Second, as the locational data are based on real-world coordinate sys-

tems, such as latitude and longitude, Universal Transverse Mercator, or
British National Grid, any dataset can potentially be integrated with any
other dataset. Thus, for example, data on specific buildings based on points
can be integrated with demographic data from census tracts represented
as polygons, transport information can be represented as lines, and data on
heights can be represented as pixels on a raster surface. This enables com-
plex representations of a study area to be built up from multiple, apparently
incompatible sources.
Third, and perhaps most important, GIS allows the researcher to explore
the topic under study in a way that explicitly considers the impact of space
and location. This might involve using formal spatial statistical methodolo-
gies (Fotheringham et al. 2000; Maguire et al. 2005) or might simply involve
asking questions about why different places appear to behave in different
ways. The increasingly sophisticated suite of statistical tools that accompany
GIS software provide insight into the strength, rather than just the existence,

of a spatial pattern, indicating how tightly grouped or widely dispersed it is.

As correlation coefficients are to a scatter plot, so measures of the character-
istics of spatial distributions are to the visualization of those patterns, provid-

ing indexed scores of the strength of complex relationships. Spatial statistics

likely will assume increasing importance in guiding the development of GIS
as this revolution turns from visualization to more ambitious analytic pur-
suits. At the very least, GIS enables and encourages the researcher to think
carefully about the geography of the topic under study and the explanatory
power of that geography.

While GIS approaches have these advantages, they also have a num-
ber of drawbacks. First, the time it takes to create a GIS database can be
large, greatly increasing the variety of costs associated with a GIS-informed
project. Second, the use of GIS software requires that the researcher learn
certain technical skills, demanding additional time and effort. Third, and
perhaps most fundamental, there is a lack of strong geographic skills among
historians who are more used to asking questions about change over time.
This relative neglect of the geographic tradition means that even with a good
database and the technical skills to use it, a researcher still needs the concep-
tual tools to frame research questions and conduct the research in ways that
make full use of the available spatial and thematic information. The good
news in respect to this last issue is that GIS has reawakened interest in the

This content downloaded from

54.146.125.244 on Wed, 20 Nov 2024 15:17:53 UTC
All use subject to https://about.jstor.org/terms
 Introduction 457

importance of geography and space to the historian (Bodenhamer et al. 2010;

Gregory 2003). Myron Gutmann (2002: ix) puts the case well: "GIS enriches
both qualitative and quantitative approaches to history . . . enabling] stu-
dents, teachers, and researchers to think differently about the past."

Historical GIS in Urban History

The most important factor in determining the extent to which historical GIS
will become an established part of the discipline of history is the success
or otherwise that its practitioners have in delivering research that demon-
strates that GIS can and is making a direct contribution to knowledge in
the discipline. Key to this is not the fact that GIS is used but instead that
the research advances our understanding of the topic under study. The field
where arguably the most progress has been made toward this goal is urban
history. A number of reasons may be identified for this. First, urban history

has a long tradition of acknowledging geographic features, as exemplified by

the work on Milwaukee produced by Kathleen Conzen (1976) and Michael
Conzen and Kathleen Conzen (1979); Michael Katz's (1975) study of Hamil-
ton, Canada; John Kellogg's (1982) study of segregation in Lexington, Ken-
tucky; Sherry Olson's (1989) study of Montreal; and the Philadelphia Social
History Project (see Hershberg 1976, 1981). Urban studies also tend to offer
the historian a rich variety of spatially referenced sources, such as maps,
addresses, street names, electoral lists, gazetteers, and tract-level data. More
pragmatically, urban areas tend to be relatively small, reducing the size of the

spatial databases required. Nevertheless, boundary problems and controlling

the size of the population as well as the number of variables being examined
are important considerations as the researcher moves from the database con-
struction phase of an urban project into its more substantive phases.
In an early paper concerned with the development of a database on
Tokyo's urban growth, Loren Siebert (2000) provides a frank account of how
he created such a database, what he sees as its potential, and the problems he
encountered in developing it. More recent papers have moved the field for-
ward by looking at how spatially informed databases can be used to make a
contribution to knowledge in urban history. Andrew Beveridge (2002) takes
tract-level data on population, ethnicity, and other socioeconomic variables
for New York City from censuses from 1900 to 2000. He uses these to show
how the city grew over the twentieth century and how different ethnic areas

This content downloaded from

54.146.125.244 on Wed, 20 Nov 2024 15:17:53 UTC
All use subject to https://about.jstor.org/terms
458 Social Science History

developed in the larger city. His approach is primarily descriptive, deploying

detailed spatial and temporal data to present a narrative of the development
of ethnic segregation. Colin Gordon (2008) follows a similar approach in his
study of the decline of St. Louis since 1945. He draws on a wider range of
sources and makes extensive use of maps to produce a book that explores how
and explains why St. Louis experienced urban blight and decay after World
War II.

Amy Hillier (2002, 2003) follows a more targeted analytic approach,

focusing on the relationship between mortgage redlining (the process of not
giving mortgages to certain areas because of the perceived problems of their
residents) and ethnicity and poverty in Philadelphia in the 1930s. She com-
bines data on home loans for a sample of specific addresses, maps of the dif-
ferent "residential security zones" used to define redlined and other areas,
data from a 1934 residential survey, and the 1940 census. Using spatial statis-
tical techniques, she shows that areas with high African American and recent
immigrant populations were more likely to be redlined than others, but she is
also able to challenge the assumption that once areas had been redlined, their
subsequent decline was inevitable because of the problems of getting a mort-
gage. Etan Diamond and David Bodenhamer (2001) follow a somewhat simi-
lar approach in a study of white flight in 1950s Indianapolis. They take data
from two censuses, 1950 and 1960, and the locations of churches in the city
at the start and end of the 1950s and compare the changing ethnic makeup of
the city with the changing locations of churches. Their aim was to explore the

assumption that mainline Protestant churches followed the white population

in abandoning inner-city areas. They found very little evidence that churches
moved out of areas that had large and growing African American populations
but found that when they did relocate, this would typically be to suburban
areas with overwhelmingly white populations.
More recent work in this field has begun to use large databases con-
structed from individual-level information. This has been particularly driven
by the awareness that segregation and spatial concentrations of social groups
can only really be understood and seen when working with data at this level.
To this end significant work has been undertaken in Hartford, Connecti-
cut (Schlichting et al. 2006; Tuckel et al. 2007), looking at African Ameri-
can populations in the city during the Great Migration. Donald A. DeBats
(2008) also uses individual-level data in a similar way but compares two con-
trasting cities: Alexandria, Virginia, and Newport, Kentucky.

This content downloaded from

54.146.125.244 on Wed, 20 Nov 2024 15:17:53 UTC
All use subject to https://about.jstor.org/terms
 Introduction 459

These examples all use primarily quantitative, social science-style

approaches to analyze cities that are almost all American. A contrast is pro-
vided by the work of Keith Lilley and colleagues, who use GIS to explore
medieval cities in England and Wales. Their approach is based on using GIS
to integrate and explore archaeological records, surveys conducted using
the Global Positioning System (GPS), and environmental data. They use
these to shed new insights into the structure of towns in this period (Lilley
et al. 2005a, 2005c) and also to disseminate information about them to a wide
audience in an interesting and attractive way through the Mapping the Medi-
eval Townscape (Lilley et al. 2005b) and Mapping Medieval Chester (Map-
ping Medieval Chester Project 2009) websites. This work illustrates that
while quantitative approaches have shown the most progress to date, per-
haps reflecting the development of historical GIS and GIS technology more
generally, the use of GIS does not force the researcher to take a particular
approach to quantitative data. Many different types of data can be included
in a GIS, and the approach to analyzing them is very much determined by
the interests of the researcher and the available sources.

This Collection

The essays in this collection mainly reflect social science approaches to ana-
lyzing urban history. Jason A. Gilliland, Sherry H. Olson, and Danielle Gau-
vreau use GIS individual-level data from Montreal for the years 1881-1901
to explore patterns of spatial separation on four dimensions: language, reli-
gion, socioeconomic status, and age. GIS is important in this investigation
because it allows the identification of spatial aggregations at various scales
that are consistent across time, facilitating comparison and testing for the
salient level of spatial separation on those four dimensions. The work iden-
tifies "frontier areas" of diversity and high levels of growth but also reveals
an industrial city committed to a pattern of housing stock that made its own
contribution to the maintenance of patterns of spatial differentiation. Mon-
treal emerges as a city with high and consistent levels of residential segrega-
tion based on ethnicity and socioeconomic status, both most noticeable at the
micro level.

The project by Donald A. DeBats on Alexandria, Virginia, and New-

port, Kentucky, uses individual-level data to map the populations of these
two small mid-nineteenth-century cities, one commercial and one industrial,

This content downloaded from

54.146.125.244 on Wed, 20 Nov 2024 15:17:53 UTC
All use subject to https://about.jstor.org/terms
460 Social Science History

one based on slave labor and one based on immigrant labor. With the popu-
lations assigned to specific addresses, GIS allows the display of social, eco-
nomic, and political data across the cities, revealing the contrasting spatial
patterns associated with their very different political economies. The work
contrasts the extent of vacant land in the cities and their use of river frontage

and explores the differences in the extent of home ownership. It uses kernel-
density measures of the distribution of social groups to explore neighbor-
hood formation and, with individual-level political information, the political
influences of such groupings.
Mathew J. Novak and Jason A. Gilliland use GIS with a database of
retailers in the commercial city of London, Ontario, between 1844 and 1916
to track the changing distribution of businesses, particularly those in the
broad categories of food retailers and fashion retailers. But of course retail
establishments and their distribution tell us far more about a city than where

people shopped; as Novak and Gilliland point out, the retail sector of a city
defines "vital places in the public realm where people congregate and inter-
act." The work suggests that while food retailers, such as butchers, opened
shops to serve the localized needs of an expanding city, businesses dealing
with fashion, especially dry goods shops, remained committed to a pres-
ence in the city's retailing core. This concentration created a crucial mass of
options that attracted customers and facilitated comparison shopping. Those
fashion shops that located in the periphery tended to be smaller and offered
less choice than the major dry goods stores in London's central retailing
district.

Finally, Aaron Raymond demonstrates the ability of GIS to visualize

change. He uses GIS to discover the extent of topographic alteration of
Seattle between 1906 and 1930 and to show the impact of these changes on
the spatial profile of the city. The specific point of his study is the massive
effort to remove from the city's business district Denny Hill, 245 feet high
and 60 blocks in area. The city developers attacked the hill and the buildings
on it in two massive "regrades" to permit the expansion of the central busi-
ness district of this major port city. Raymond's work preserves the building
dimensions and shapes that were lost over a 105-year period of Seattle's his-
tory, allowing a unique view of urban development over time.

This content downloaded from

54.146.125.244 on Wed, 20 Nov 2024 15:17:53 UTC
All use subject to https://about.jstor.org/terms
 Introduction 461

Conclusion

Each of these articles shows how GIS can make a contribution to our under-

standing of urban history. Most of them do this with large datasets about
individual people, households, or properties. They then focus on a range of
questions that stress the geographic aspects of urban history, including segre-

gation, core and periphery, and topography. The GIS provides the framework
that helps the researcher ask questions concerning what, where, and when.
The final and most important stage in the research process is to ask why. GIS
does not of itself answer this. Instead, it provides the descriptive informa-
tion that the researcher has to explain. In this way the GIS enhances analytic
skills through its ability to summarize large amounts of complex information
in space and time. Explaining why these patterns are as they are remains the
task for the skilled historian, not the computer. It is well known that learn-
ing GIS skills and building GIS databases represents a major investment of
time and effort. Each article describes this effort in some detail. The ques-
tion remains, is it worth it? Judging by the innovativeness and success of the

articles in this collection, the answer is clear- yes, it is. In all of this, there are

clear signs that GIS is encouraging a revival of urban history.

References

Beveridge, A. A. (2002) "Immigration, ethnicity, and race in metropolitan New York,

1900-2000," in A. K. Knowles (ed.) Past Time, Past Place: GIS for History. Red-
lands, CA: ESRI Press: 65-78.
Bodenhamer, D. J., J. Corrigan, and T. M. Harris, eds. (2010) The Spatial Humanities:
GIS and the Future of Humanities Scholarship. Bloomington: Indiana University
Press.

Conzen, K. N. (1976) Immigrant Milwaukee, 1836-1860: Accommodation and Commu-

nity in a Mid-Nineteenth-Century City. Cambridge, MA: Harvard University Press.
Conzen, M. P., and K. N. Conzen (1979) "Geographical structure in nineteenth century
retailing: Milwaukee, 1836-90." Journal of Historical Geography 5 (1): 45-66.
DeBats, D. A. (2008) "A tale of two cities: Using tax records to develop GIS files for map-
ping and understanding nineteenth-century U.S. cities." Historical Methods 41 (1):
17-38.

Diamond, E., and D. Bodenhamer (2001) "Investigating white-flight in Indianapolis: A

GIS approach." History and Computing 13 (1): 25-44.
Fotheringham, A. S., C. Brunsdon, and M. Charlton (2000) Quantitative Geography:
Perspectives on Spatial Data Analysis. London: Sage.

This content downloaded from

54.146.125.244 on Wed, 20 Nov 2024 15:17:53 UTC
All use subject to https://about.jstor.org/terms
462 Social Science History

Gordon, С. (2008) Mapping Decline: St. Louis and the Fate of the American City. Phila-
delphia: University of Pennsylvania Press.

Gregory, I. N. (2003) A Place in History: A Guide to Using GIS in Historical Research.

Oxford: Oxbow Books.

Gregory, I. N., and P. S. Ell (2007) Historical GIS: Technologies, Methodologies, and
Scholarship. Cambridge: Cambridge University Press.
Guttmann, M. (2002) "Preface," in A. K. Knowles (ed.) Past Time, Past Place: GIS for
History. Redlands, CA: ESRI Press: vii-x.
Hershberg, T. (1976) "The historical study of urban space: Introduction." The Philadel-
phia Social History Project (special issue). Historical Methods Newsletter 9 (2-3):
99-134.

Hershberg, T., ed. (1981) Philadelphia: Work, Space, Family, and Group Experience in
the Nineteenth Century; Essays toward an Interdisciplinary History of the City.
New York: Oxford University Press.

Hillier, A. E. (2002) "Redlining in Philadelphia," in A. K. Knowles (ed.) Past Time, Past

Place: GIS for History. Redlands, CA: ESRI Press: 79-93.

Journal of Housing Research 14 (1): 137-68.

Katz, M. (1975) The People of Hamilton, Canada West: Family and Class in a Mid-
Nineteenth-Century City. Cambridge, MA: Harvard University Press.
Kellogg, J. (1982) "The formation of black residential areas in Lexington, Kentucky,
1865-1887." Journal of Southern History 48 (1): 21-52.
Knowles, A. K. (2000) "Introduction." Social Science History 24 (3): 451-70.

Spatial Data, and GIS Are Changing Historical Scholarship. Redlands, CA: ESRI
Press: 1-26.

Lilley, K., C. Lloyd, and S. Trick (2005a) "Mapping medieval urban landscapes: The
design and planning of Edward I's new towns of England and Wales." Antiquity
(303): www.antiquity.ac.uk/projgall/lilley/index.html (accessed October 20, 2009).

Edward I, ads.ahds.ac.uk/catalogue/specColl/atlas_ahrb_2005/index.cfm (accessed

October 20, 2009).
Lilley, K., C. Lloyd, S. Trick, and C. Graham (2005c) "Mapping and analysing medieval
built form using GPS and GIS." Urban Morphology 9 (1): 5-15.
Maguire, M., M. F. Goodchild, and M. Batty, eds. (2005) GIS, Spatial Analysis, and
Modeling. Redlands, CA: ESRI Press.
Mapping Medieval Chester Project (2009) Mapping Medieval Chester: Place and Identity
in an English Borderland City, c. 1200-1500, www.medievalchester.ac.uk (accessed
July 25, 2011).

Olson, S. (1989) "Occupations and residential spaces in nineteenth-century Montreal."

Historical Methods (22): 81-96.
Schlichting, K., P. Tuckel, and R. Maisel (2006) "Residential segregation and the begin-

This content downloaded from

54.146.125.244 on Wed, 20 Nov 2024 15:17:53 UTC
All use subject to https://about.jstor.org/terms
 Introduction 463

ning of the Great Migration of African Americans to Hartford, Connecticut: A GIS-

based analysis." Historical Methods 39 (3): 132-43.
Siebert, L. (2000) "Using GIS to document, visualize, and interpret Tokyo's spatial his-
tory." Social Science History 24 (3): 537-74.
Tuckel, P., K. Schlichting, and R. Maisel (2007) "Social, economic, and residential diver-
sity within Hartford's African American community at the beginning of the Great
Migration." Journal of Black Studies 37 (5): 710-36.

This content downloaded from

54.146.125.244 on Wed, 20 Nov 2024 15:17:53 UTC
All use subject to https://about.jstor.org/terms