GIS and quantitative geography

Author(s): Keiji Yano

Source: GeoJournal , 2000, Vol. 52, No. 3, The Contribution of GIS to Geographical
Research (2000), pp. 173-180
Published by: Springer

Stable URL: https://www.jstor.org/stable/41147551

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

Springer is collaborating with JSTOR to digitize, preserve and extend access to GeoJournal

This content downloaded from

134.226.213.129 on Fri, 01 Nov 2024 14:04:19 UTC
All use subject to https://about.jstor.org/terms
 ^* GeoJournal 52: 173-180,2001. ^
■p™ © 2001 Kluwer Academic Publishers. Printed in the Netherlands.

GIS and quantitative geography

Keiji Yano
Department of Geography, Ritsumeikan University, Kyoto, Japan (E-mail: yano@lt.ritsumei.ac.jp)

Received 31 March 2001; accepted 5 October 2001

Key words: applied geography, GIS, GIS revolution, Japanese human geography, quantitative geography

Abstract

The purpose of this paper is to explore the development of the GIS revolution within the field of geography in the 1980s and
1990s, taking into consideration the influence of external and internal factors on disciplinary change. Geography is rapidly
changing by the impact of external factors on geographical research and also as a consequence of economic recession and
cutbacks in higher education. Most geography departments in universities have tried to approach this impact by promoting
or shifting to human geography as an applied science, and by offering relevant skills through GIS as an approach to
contemporary problems. It is concluded that quantitative geography is essential for the further expansion of GIS within
geography and also for the survival of geography.

Introduction geography would be affected by the trends in the US and

UK.

The GIS revolution starting from the late 1980s has changed
from 'GISystems' to 'GIScience' (Goodchild, 1992). This
Before the GIS revolution
paper explores the development of the GIS revolution within
geography in the 1980s and 1990s, considering the influence
of external and internal factors on disciplinary change. This section will evaluate various debates and criticisms
about quantitative geography and their effect on the GIS
GIS is an interdisciplinary field in which research fields
including geography have traditionally employed a spatialrevolution, based on the results shown in Johnston (1997).
perspective. Moreover, GIS is considered as a big industryIn the beginning of the GIS revolution, the attitude of
or business in the field of applied sciences. It has much anti-quantitative geographers was seen to be as follows. The
social relevance considering that it is applied in various epistemology
ar- and method that underpins GIS emerged under
eas such as marketing GIS, GIS for welfare, GIS for local the auspices of positivist and empiricist versions of science,
government, and GIS for disaster prevention. Therefore, and it reemerged as a result of the collaboration of quantitative
is insufficient to recognize the GIS revolution as academicmethods and computer developments of the 1970s. Taylor
development. The process of GIS revolution will be made (1990) skeptically criticizes the GIS revolution; 'GIS is ge-
clear through the interaction between discipline and society.
ography's own little bit of the ' high-tech ' revolution and has
This paper attempts to focus on the interrelationshipsuffered
be- accordingly with the seemingly endemic high-tech
tween GIS and quantitative geography, and the question disease of mega-hype' Taylor (1990) suggests that GIS en-
of how GIS emerged out of intellectual trends within ge- thusiasts base their research on a naive empiricist approach
ography. Since the 1970s, quantitative geography, thatand is, have failed to transform GIS into GKS (Geographi-
positivist spatial science, has been under considerable attack
cal Knowledge Systems); emphasizing also that 'they (GIS)
within the discipline of geography in the US and UK. Itwill has leave geography intellectually sterile - high-tech trivial
also not been the main stream in human geography. Yet,pursuit'
the .
GIS revolution has been promoted by geographers who have Goodchild (1990), who was one of the key-persons in
continued with their research in quantitative geographythe inNCGIA (National Center for Geographic Information
the 1970s and 1980s. It is seen that many departments ofand ge- Analysis), responded by stating that GIS research in
ography in the US and UK have attempted to actively adopt geography is concerned not with technology, but with the
GIS into their education system, with the aim to improve issues
the surrounding its use. He further pointed out that a ma-
status and quality of geographical studies. These trends have
jor change in the technology of mapping might be expected
prompted a shift towards applied geography. This partially to provoke reasonably profound thoughts from geographers.
seems to be a response to changes occurring in the social Openshaw's (1991) response was also aggressive, arguing
contexts within which human geographers work, depending that there can be no knowledge without information but also
also on the research funding. The experiences evident claiming
in that GIS could resolve the identity crisis within
the US and UK suggest that the future of Japanese human

This content downloaded from

134.226.213.129 on Fri, 01 Nov 2024 14:04:19 UTC
All use subject to https://about.jstor.org/terms
174

geography. This was followed by a few more replies, for work and provide information pragmatically. Furthermore,
example, Taylor and Overtoil (1991) and Openshaw (1992). he proposes that data-driven computer modeling in an in-
According to Johnston (1997, p. 315), this debate has formation economy such as computational geography is a
brought out some extreme views that clarify many of the means of ensuring the survival of geography as an academic
differences between 'social theory geographers' and 'GIS- discipline.
quantifiers'. Many amongst the former doubt the value of In the case of quantitative analysis, Cox (1989) claims
that the advanced analytical procedures are not generally
quantitative work, irrespective of whether it was done using
GIS or not. Their concerns are about the use of GIS in so- available and that the procedures have not been applied to
ciety (Pickles (ed.), 1995; Curry, 1998). Meanwhile, spatial
problems that are in the mainstream. As a result, adherents
analysis will no doubt continue to attract geographers,toboth data based approach within GIS have been able to promote
for itself and for its potential applied value. more readily appreciated expertise with a largely empiricist
Recognizing GIS as the successor of quantitative geog- orientation.
raphy, it is necessary to understand the relationship between The 1980s experienced some negative impacts on quan-
GIS and quantitative geography again. Taylor and Johnston titative studies in geography above mentioned. Quantitative
(1995) identify two tensions facing the GIS revolution,geography
i.e., itself had two sides, i.e. of deductive and in-
one between inductive and deductive 'sciences', and the ductive, and academic and applied. The GIS revolution has
other between 'pure' and 'applied' geography. progressed under such a situation.

Debate on quantitative geography Inductive and deductive routes in quantitative geography

After the quantitative revolution, there has been an emer- After the origin of quantitative revolution in the late 1950s
gence of new paths and principles that has transformed the from the US, quantitative geography had shifted from statis-
central questions and approaches in the discipline. Quantita- tical to theoretical geography (Wilson, 1972). Quantitative
tive geography had not been the main stream within geogra-geography developed in a group led by the Washington
phy in the 1970s and 1980s, and has been under considerable school, and is mainly descriptive statistical geography. This
attack among anti-quantitative geographers. However, thisapproach may be divided into multivariate analysis for ge-
attack has had little effect on the volume of work done ographical matrices, i.e. regression and factor analysis, and
by quantitative geographers. The book Quantitative Geog- spatial pattern analysis for spatial elements. The latter has
raphy (Wrigley and Bennett (eds), 1981) has significantlybeen taken over by the Bristol school and has expanded
to spatio-temporal modeling. Statistical geography in its
contributed to the field of human geography. This endeavor
certainly connected to the GIS revolution that has been earlier
stim- days suffered from the inadequate use of borrowed
ulated and aided by advances in data handling and displaytechniques, which were developed for aspatial contexts; for
technologies. example, the problems of spatial autocorrelation. As a re-
Remodeling Geography (Macmillan (ed), 1989) thatsult, wasthe use of statistical methods is accepted as descriptive
published for the twentieth anniversary of the publication
statistics
of and not as inferential statistics that deals with popu-
Chorley and Haggett's (1967) seminal work Models inlation,
Ge- random sampling, and so on. To overcome the various
problems related to statistical assumptions, advanced spa-
ography, focuses on the debate about quantitative geography
tial statistical models, i.e., spatial time series models and
while also rethinking the meanings of models in geography.
categorical
Harvey (1989) claimed that accepting the advance of data analysis, i.e., a logit model, have been
modeling research and pursuing knowledge using the developed.
pos-
Today, GIS incorporates standard tools in spatial analy-
itivist route did not necessarily generate usable configu-
sis, ranging from the point pattern analysis to spatio-
rations of concepts and theories. This was with reference
to examples such as the urban modeling by Wilson
temporal
and modeling (Fotheringham and Rogerson (eds),
1994). Advanced spatial statistics has made rapid progress
spatio-temporal series modeling by Haggett in the 1970s.
Although Wilson (1989) and Macmillan (1989) accepted theto parallel computing and artificial intelligence in the
related
GIS environment (Fisher and Getis (eds), 1997).
limitations of modeling research, they claimed that models
There was a great advance within theoretical geography
were important tools that could help researchers to handle
when Alan G. Wilson developed the entropy-maximizing
complexity in a variety of situations. Furthermore, model-
models and urban and regional modeling with his colleagues
ing also involves the development of theories that contain
quantifiable propositions. at Leeds University in the 1970s (Yano, 1990). Following
this, an interesting ability has been observed in spatial in-
There are a few different types of quantitative geog-
raphers. Openshaw (1989) states that urban and regional
teraction models and structures to reproduce the results of
modeling based on a hypothetic-deductive framework, classical
in location theories (Birkin and Wilson, 1986a, b).
relation to the understanding and application of work,So far GIS has not had much to do with theoretical geog-
has
been taken too far and has not been particularly successful.
raphy and has been mainly limited to the mapping of results.
He accepts there is no doubt about either the quality However,
of the there is a potentially fruitful connection between
work or the utility of the mathematical models, ratherGIS his
and spatial modeling. The use of GIS in location-
criticism is focused on the lack of applied relevancy. He modeling for the Spatial Decision Support System
allocation
tries to add the value to geographical data with inductive
gives validity and efficiency to the models (Densham, 1996).

This content downloaded from

134.226.213.129 on Fri, 01 Nov 2024 14:04:19 UTC
All use subject to https://about.jstor.org/terms
 175

Moreover, the application of new techniques such as parallel Claims that geographers and their research should be
computing and artificial intelligence has led to an improve- more relevant to society are not new. The development
ment in processing capability, and has stimulated progress in of social relevancy in geography after the Second World
the GIS industry and business together with its shift towards War has originated from a formal objection of the scientific
applied geography (as will be shown later in Birkin et al. worldview, as well as the various social problems such as
(eds), 1995; Birkin, 1996). depression, social inequality, environment pollutions and so
on, that emerged after the failure of the economic boom.
Shift to applied geography This extension was identified as radical geography. How-
ever, apart from the above trend, a lot of contributions have
In the 1990s, quantitative geography tied up with GIS using been made by applied geographers to society, both in the
a descriptive, inductive and applied view. This trend was field of research and teaching. British geography has a strong
supported by the rapid development of computer technol- traditional link with the broad discipline of planning. It orig-
ogy and the huge data storage of geographical information. inated Stamp's Land Utilization Survey of Britain in the
Moreover, it is very important to consider the interaction 1930s. After the Second World War, landuse planning was
between geography and social relevancy. Most geography established on a large scale, which was followed by urban
departments in universities in the US and UK have attempted modeling providing resources for the joint activities between
to promote human geography as an applied science by of- landuse and transport planning (Wilson, 1974; Batty, 1976;
fering relevant skills using GIS. This shift has contributed Yano, 1990). The potential for geographic inputs on spatial
towards the award of research funding and the recruitment of planning was promoted by the allocation of landuse and traf-
undergraduate students with interest in geography (Johnston, fic flows. Most of this was simple pragmatic application of
1997). technical skills, although there were also some attempts to
The relationship between academic geography and its so- evaluate policy impacts and to develop a theory of decision-
cial context cannot be ignored. For example, the shift from making in this context. This has resulted in stronger links
regional geography to systematic geography in the quanti- among academics, local governments and planning con-
tative revolution was pressed by an evaluation of geography sultants, as well as the tradition of policy-related research
based on its contribution to society. geared to service planning and resource allocation.
In the UK, there has been a radical change in the situ- Most of this work is set in an empiricist and usually
ation of universities and disciplines under the retrenchment implicit positivist framework (Johnston, 1997). Academic
policy of the Thatcher and Major administrations, introduc- geography has shifted to applied geography that has been
ing thereby a market mechanism to the competition among involved in the GIS revolution. The trend has also been
universities and their evaluation. The new evaluation system enhanced by the institutionalization of GIS since the late
has especially affected the direction of academic research. 1980s.
After 1986, the University Grants Committee produced a
rating of the research record and began a selective allocation
of funds to universities that reflected positive rating. In the Institutionalization of GIS
research assessment undertaken for the fourth time in 1996,
research money was allocated to each university based on Experiences in the US
the criterion of not only the quality of research, but also on
their potential to contribute to society. This evaluation also In 1988, the United States National Science Foundation an-
took into consideration the quality of education before con- nounced the award of a grant of 1 . 1 million dollar per year
firming a reasonable allocation of funds. The allocation of for a period of five years, to establish the NCGIA. This grant
public money as funds to each university was also done very was awarded to a consortium of universities including of
selectively based on the number of students and the nature the University of California at Santa Barbara, the University
of the research and education at the universities. Therefore it of Maine and the State University of New York at Buffalo.
The aims of establishments of these centers are described as
is seen that each department attempts to promote their status
over the rating. follows: (1) to advance the theory, methods, and techniques
The necessity of an applied aspect within human ge-of geographic analysis based on geographic information sys-
ography has increased in the past few decades. In periods tems (GIS) in the many disciplines involved in GIS research;
of economic recession, cutbacks are expected in the pub- (2) to augment the nation's supply of experts in GIS and
lic funding towards higher education and research. In such geographic analysis in participating disciplines; (3) to pro-
cases, geographers as well as researchers in other fields are mote the diffusion of analysis based on GIS throughout the
forced to seek financial support elsewhere. Geographers par-scientific community; and (4) to provide a central clearing-
ticularly have to continue to prove the value of geographyhouse and conduit for disseminating information regarding
to outsiders, while also providing insiders with a strong in-research, teaching, and applications.
tellectual foundation. These can be considered as external Thus, the anticipated mission of knowledge is for a wider
influences on disciplinary developments. It can be said that academic and practitioner communities (Fotheringham and
both pure and academic geography are necessary for MacKinnon,
the 1989).
future of this discipline (Johnston, 1997). The fact that NSF had a will to support geographic
research inspired many universities to look to their own

This content downloaded from

134.226.213.129 on Fri, 01 Nov 2024 14:04:19 UTC
All use subject to https://about.jstor.org/terms
176

geography departments for potential academic growth. Sim- school has long been associated with quantitative geogra-
ilarly, the assignment of NCGIA to geography has created phy. Under the leadership of Alan Wilson, it has established
new links between major computing hardware and software a reputation in the fields of urban and regional modeling
companies with geography departments inside and outside through the application of Wilson's entropy maximizing spa-
the NCGIA. tial interaction models. Throughout the 1980s, much of
this work was theoretical in nature and was used to illus-
In the American background of the relationship between
trate the complexities of urban systems. It was also used
geography and GIS, factors such as the external environment
of the discipline; economic depression after the late 1970s,
to demonstrate the ability of models of spatial interaction
and structure to reproduce the results of a few classical lo-
the reduction of budget for higher education, and cut of fund-
ing for social sciences influenced the tendency of academic
cation theories as mentioned before. Although these works
geography. The crisis within geography was symbolized contributed
by to the development of substantive geographic
theories,
the close of the geography department at the University of it was a recognized fact that there was a gap
Michigan. This promoted the strategy of AAG to empha- between them and empirical research.
size the importance of applied geography, recognizing alsoThe mid-1980s however, saw a major shift at Leeds.
the need to attract students who are more conscious of their There was more emphasis on theory and application that
careers and skills (Johnston, 1997). were driven by new theoretical developments (Clarke et al,
1995). This timing corresponded with not only the beginning
Experiences in the UK of the GIS revolution, but also the pressures on univer-
sity academics to obtain more external financial support
GIS and geographical information processing obtained pub-for research activities. For the Leeds School, the arrival
lic recognition in the UK as being essential and up-to-date
of GIS was seen as producing a new opportunity for the
research tools, during the university reform movement by
development of such applied spatial modeling.
the Thatcher and Major administrations. This had a signif-
Considering that the Leeds school was not amongst the
icant influence on academic geography. A key landmark in
subsequent beneficiaries of the RRL program, it was driven
the development of GIS in the UK was the publication of
rather by necessity for getting other funding. GMAP Limited
the Chorley Report in 1987. Its recommendations were, the
was founded in the University of Leeds School of Geog-
establishment of a center for advice on the handling of geo-
raphy in 1989 by A.G. Wilson, M. Clarke and M. Birkin.
graphical data in order to service the wider GIS community,
After Wilson took office as the Vice Chancellor of Leeds
and the encouragement of academics to actively disseminate
University, Stan Openshaw took over the post of profes-
information on the applications and benefits of GIS. The
sor as his successor. Openshaw advocated computational
UK's Economic and Social Research Council (ESRC) was to
geography aided by artificial intelligence and parallel com-
announce a big foundation for GIS research to underpin the
puting. Unfortunately however, he retired due to an illness.
Regional Research Laboratories (RRL) which were located
Subsequently, Professor D. Dorling, who was famous for
in university departments of geography such as the NCGIA
cartogram mapping, moved to Leeds University in his place.
in the US.
In 1994, GMAP employed 75 full-time staff includ-
The main objectives of the RRL Initiative were, to indi-
ing graduate geographers with skills in spatial analysis and
cate the need to establish a resource base for research and
modeling, business administration, computer programming.
policy analysis, to highlight the need for work on method-
It had an annual revenue of 3.5 million pounds (Clarke
ological issues relating to the management of large-scale
and Clarke, 1995). Business at GMAP includes market
data resources and finally, to recognize the need to develop
modeling, optimization of retailer location and network,
local centers of expertise capable of meeting the distinc-
maximization of revenues and profit and so on. This com-
tive requirements of different regional research communities
pany attempted to integrate GIS technology with geograph-
within the UK. Within the trial phase of the RRL Initiative
ical modeling techniques in order to contribute to a better
running from February 1987 to October 1988, four RRL
understanding and planning in both public and private sec-
centers were established at Scotland, North England, Wales
tors. During the 1970s and 1980s, retailers' investment was
and south west and south east England. The objectives of
largely associated with network expansion for increasing
these centers included the construction and management of
market shares. The recessionary climate of the 1990s also
regional databases, the development of analytical systems
did not necessarily encourage such expansionary strategies.
and the diffusion and education of GIS researches. The ad-
The battle for market share or profitability then shifted to
dition of four more centers, i.e., North Ireland, North East
the search for greater returns from the existing store network
England, Midland, and Liverpool and Manchester, began in
(Clarke and Clarke, 1995).
October 1988 and continuing December 1991 with over 2 GMAP develops customized spatial decision support
million pounds fund allocated by the ESRC. The ESRC software
has for a wide variety of blue-chip clients in Europe
since provided limited funds for the continued networking
and North American. Its clients include a variety of retail
of the RRLs (Clarke et al, 1995).
and commercial sponsors, in contrast to those of the RRL
Apart from the RRL, there was a precedence of coop- who are mainly public sectors agencies. If the RRL Ini-
eration with the private sector involving GIS in its various
tiative represents a coordinated 'top-down' approach in the
forms, as done by the GMAP company established by interaction
the between academia and business, the Leeds story
school of geography at the University of Leeds. The Leeds

This content downloaded from

134.226.213.129 on Fri, 01 Nov 2024 14:04:19 UTC
All use subject to https://about.jstor.org/terms
 177

on the other hand represents a 'bottom-up' response from Key Centre for Social Applications of Geographical Infor-
one university department towards the demand for applied mation Systems (GISCA) at University of Adelaide, Aus-
consultancy within particular niches of the fast developing tralia, and, Geographic Visualization Science, Technology,
GIS research area. GMAP has since expanded its business and Applications Center (Geo VISTA) at the Pennsylvania
into American and European markets through collaboration State University. In Japan also, the Center for Spatial Infor-
with other agencies and institutions (Clarke et al, 1995). mation Science (CSIS) was established at Tokyo University
It is difficult to appreciate the quantity of GMAP's work in 1998.

that has been done in quantitative geography, because un- There have also been publications of several periodi-
like pure academic research, little of it is reported in the cal journals devoted to GIS. For example, the International
discipline's academic literature. This is in part due to the Journal of Geographical Information System founded in
confidentiality constraints imposed by clients. However, a 1987 (in 1997, the title was changed to the International
glimpse of several examples of work done so far is available Journal of Geographical Information Sciences) Transac-
in Longley and Clarke (1995) and Birkin et al (1995). tions in GIS (in 1996), Geoinformatica (in 1997), Geograph-
The very efficient partnership between academia and ical & Environmental Modelling (in 1997) as an academic
business built in the UK, such as that in academia geode- journal, and GIS World (in 1988) and Mapping Awareness
mographics and SDSS, are theoretically developed through (in 1987) as a commercial journal. In 1993, the GIS associa-
the RRL. At GMAP on the other hand, these systems are tion in Japan issued Theory and Applications of GIS (Okabe,
applied and verified to data in the real world, and the results 1999). There has also been rapid consolidation of the GIS
are effectively used by their client agencies. Any profit ob- software in Japan in the 1990s, followed by the publication
tained by academic geographers from such client agencies is of some introductory books. This has greatly promoted the
finally returned to the research fund. Such a reinforcement of popularization of GIS in Japan (Yano, 1999).
applied geography with social relevancy will thus contribute
to getting more funding from research councils.
New quantitative geography under the GIS environment
Expansion in the 1990s
The GIS revolution was already over in the first half of the
The institutionalization of academics means that the acad- 1990s. The relationship between GIS and spatial analysis
has since begun to be investigated. The initiative on GIS
emic area or field becomes a profession. In the first place,
although the boom of GIS was indeed influenced by progressand spatial analysis submitted to the NCGIA in March 1989
of computer technology, the development of GIS was also was formally resubmitted in June 1991 after being divided
into separate proposals, i.e., initiatives in GIS and statistical
maintained by institutionalization at various levels. As men-
tioned before, the United States NSF had invested 5.5 analysis, and, GIS and spatial modeling. The essence of the
million dollars into the NCGIA, based at three departmentsformer was accepted and evolved into the more generic 'GIS
for a period of 5 years starting in 1988. The British ESRCand Spatial Analysis' initiative, with the expectation that an
moreover had established a network of RRL with eight sitesinitiative to emphasize spatial modeling would take place at
from 1987 through 1991, including a one-year trial phase. a later date (Rogerson and Fotheringham, 1994). The estab-
As a result, it can be said that the GIS revolution was driven
lishment stage of spatial data and the popularization of GIS
forward by the NCGIA and the RRL. have thus led to the start of a research program for pattern
Above all, 21 research initiatives were carried at NC-analysis of spatial data and their interpretation.
GIA since 1988, with only a few cancellations or changes. It is difficult to define spatial analysis when considering
its relationship with GIS. Bailey's (1994) broad definition
Among these initiatives, the following ones provide impor-
tant themes showing a close relation between geography on spatial analysis indicated it to be a general ability to
and GIS; Accuracy of Spatial Databases (1998-1990; Good- manipulate spatial data into different forms and exact addi-
child M.), Spatial Decision Support Systems (1990-1992; tional meaning as a result. Methods addressing the inherent
Goodchild M. and Densham P.), Spatio-Temporal Reasoningstochastic nature of patterns and relationships are called sta-
in GIS (1993-; Egenhofer M. and Golledge R.), GIS and tistical spatial analysis, rather than forms of analysis that are
Spatial Analysis (1992-1994; Fotheringham S. and Roger- purely deterministic, and are divided into the spatial sum-
son P.) and The Social Implications of How People, Space marization and the spatial analysis of such data. The former
and Environment are Represented in GIS (1995-; Harris refers to the basic functions for a selective retrieval of spatial
T. and Weiner D.). Following this, the NCGIA continued information within defined areas of interest, and for the com-
with further three initiatives such as the Varenius project
putation, tabulation or mapping of various basic summary
beginning in 1997: cognitive models of geographic space; statistics for this information. The latter is more concerned
computational implementations of geographic concepts; and
with the investigation of patterns in spatial data - in partic-
Geographies of information society. ular, in seeking possible relationships between such patterns
In the 1990s, some institutions were established in co-and other attributes or features within the study region, and
operation with NCGIA and RRL, for the promotion of GIS also with the modeling of such relationships for the purpose
research. These were the European Umbrella Organisationof understanding or prediction (Bailey, 1994).
for Geographic Information (EUROGI) in Europe, Centre Such spatial analysis has required complex programs un-
for Advanced Spatial Analysis (CASA) at London, Nationaltil now and few extreme researchers have been able to use

This content downloaded from

134.226.213.129 on Fri, 01 Nov 2024 14:04:19 UTC
All use subject to https://about.jstor.org/terms
178

them. Nowadays however, the end users of spatial analysis geographers in the 1960s, the quantitative revolution in
increase through the extensions and links between spatial Japanese geography occurred later in the first half of the
analysis and statistical package software, or the inset in GIS 1970s. This development was unique and different in two
software. Together with the popularization of traditional spa- ways from the experience of Western countries, as suggested
tial analysis, there has also been an appearance of various by Johnston (1997). The first is that, Japanese quantita-
huge spatial data forces, as well as the development of new tive geography selectively accepted a part of research topics
methods and models in spatial analysis (Fotheringham et al, related to quantitative geography in the US and UK, es-
2000). Under the GIS environment, new research topics are pecially the descriptive statistical geography based on the
created that could retrieve or visualize geographically mean- multivariate analysis. The second is that, Japanese geog-
ingful data from database on a cyber space, such as data raphy imported quantitative geography from overseas and
mining and pattern hunting (Openshaw and Abrahart (eds), simultaneously faced heated criticisms related to this field
2000; Hearnshaw and Unwin (eds), 1994; Dorling, 1995). in the 1970s.
The initiative of the NCGIA, GIS and Spatial Analysis, is A uniqueness of Japanese quantitative geography is that
closely related to quantitative geography, and was success- it is traditional geography that describes a region using quan-
fully completed in 1995. Following this, the key persons of titative and statistical methods. It was the point of contact
the initiative, A. Stewart Fotheringham and Michael Batty, between the traditional geography and new geography re-
have moved to the Department of Newcastle University and lated to in regional concepts, formal regions and functional
the CASA at the University of London, respectively. The regions. As a result, the main stream of Japanese quan-
main stream of quantitative geography aided by GIS is thus titative geography was based on multivariate analysis and
seen to be shifting to the UK. represented by factors such as regional division and areal
In the 1990s, there was an occurrence of the alternation classification, central-place theory (Morikawa, 1980) and
of generations in the UK tradition of quantitative geogra- factorial ecology. After the late 1970s, descriptive quanti-
phy. Peter Haggett of Bristol University, one of the leaders tative methods have been applied, although these research
in quantitative geography, has retired from the front line topics were taken in urban system research (Tanabe (ed.),
(Cliff et al (eds), 1995). The name of the school was later 1982). These approaches, which summarize the geograph-
changed from geography to geographical sciences, with Paul ical data matrices have become popular supported by the
Longley as the head. This school explores geocomputation package software SAS and SPSS (Murayama, 1990). How-
as a new tool within quantitative geography (Longley et al, ever, the approach was severely criticized for the reason that
1998), although Longley moved to a new institution recently. results were predictable and not exceptional.
Similarly, Leeds University is advocating computational ge- The advanced spatial statistics and theoretical quanti-
ography in close cooperation with the GMAP as mentioned tative geography developed in the 1970s and 1980s, have
before. This is being done by a group with Stan Openshaw not efficiently spread over Japanese quantitative geography.
as the head (though he retired now as mentioned before), Rather, they have been applied in fields such as social engi-
making use of a huge quantity of geographical information neering, urban planning, urban economics, regional science
and assisted by the latest artificial intelligent technology and civil engineering that have social relevancy. This situ-
(Openshaw and Openshaw, 1997). The contribution to the ation has been brought about by the fact that geography at
private and public sectors through GIS, for instance, in Japanese universities has contributed more to general educa-
the form of census geography (Openshaw (ed.), 1995) and tion rather than applied activities such as urban and national
geodemographics, is now greatly related to the survival of planning and business. There have been few research ac-
geography. tivities in applied geography using GIS, with the exception
Only thirty years have passed since the original 'Spatial of the disaster prevention research after the Hanshin-Awaji
analysis' was edited by Berry and Marble (1968). A book earthquake. In recent years, it is also expected that acad-
with the same title was published in 1996 by Longley and emic geography should enter applied research based on GIS,
Batty, whose subtitle was changed from 'a reader in statisti- collaborated with public and private sectors.
cal geography' to 'modelling in a GIS environment'. Such However, the development of GIS within geography, es-
a new trend can no longer stay within the framework of pecially human geography, is not anticipated as it stands.
geography. The new science of spatial information science This is because expansion of GIS in the US and UK for the
is fast growing with a tie-up with the business world. last ten years has been maintained by quantitative geography
that progressed in the 1970s and 1980s. Such an advanced
form of quantitative geography has not sufficiently devel-
GIS and Japanese quantitative geography oped in Japanese human geography. Moreover, Japanese
geography is only distantly connected with planning and
This section will consider the relationship between geogra- business, and it is expected that this linkage not come up to
phy and GIS in Japan. The tendency of Japanese quantitative our expectations in the future. In the US, there is relatively
geography has not followed that in the United State nor in little interaction between business, academic geography and
the United Kingdom. According to a survey of some ma- GIS, because of the lack of identification of geography ei-
jor Japanese journals (Yamada, 1986; Ishikawa, 1994), it ther as a profession or as a discipline in the general public
is found that although the development of quantitative ge- (Sherwood, 1995). In order to enhance the cooperation be-
ography in western countries was introduced to Japanese

This content downloaded from

134.226.213.129 on Fri, 01 Nov 2024 14:04:19 UTC
All use subject to https://about.jstor.org/terms
 179

tween geography and business, it is necessary to assert that Birkin M., 1996: Retail Location Modelling in GIS. In: Longley P.A.
and Batty M. (eds), Spatial Analysis: Modelling in a GIS Environment,
geography is useful as an applied science with GIS in the
pp. 207-226. Geoinformation International, Cambridge.
area of geographical education. Birkin M. and Wilson A.G., 1986a: Industrial Location Models 1: A Review
There is some apprehension that the acceptance of GIS and Integrating Framework. Environment and Planning A 18: 175-206.
in Japanese geography will be limited only to the use of Birkin M. and Wilson A.G., 1986b: Industrial Location Models 2: We-
ber, Palander, Hotelling and Extensions within a New Framework.
GIS as a mapping tool, especially if Japanese geography
Environment and Planning A 18: 293-306.
does not put more effort to adapt new quantitative geogra- Birkin M., Clarke G.P., Clarke M. and Wilson A.G. (eds), 1995: Intelligent
phy and to shift to applied geography. This new quantitative GIS: Location Decision and Strategic Planning. Longman, Cambridge.
geography in the GIS environment advancing in western Chorley R.J. and Haggett P. (eds), 1967: Models in Geography. Methuen,
London.
countries needs to go beyond plain geography. In fact, it is
Clarke G. and Clarke M., 1995: The Development and Benefits of Cus-
seen that a new trend of GIS and quantitative geography is tomized Spatial Decision Support Systems. In: Longley P. and Clarke G.
turning away from the main stream of geography in Japan, (eds), GIS for Business and Service Planning, pp. 227-245. Geoinfor-
and thereby creating new fields such as spatial information mation International, London.
Clarke G., Longley P. and Masser L, 1995: Business, Geography and Acad-
science and geocomputation. Such a process will promote
emia in the UK. In: Longley P. and Clarke G. (eds), GIS for Business and
further restructuring of human geography in Japan. Service Planning, pp. 271-283. Geoinformation International, London.
Therefore, it is important that Japanese geographers Cliff A.D., Gould PR., Hoare A.G. and Thrift N.J. (eds), 1995: Diffusing
should show much interest in new quantitative geography, Geography: Essays for Peter Haggett. Blackwell, Oxford.
Cox N. J., 1989: Modelling, Data Analysis and Pygmalion's Problem.
and hold a debate on this trend. Moreover, it is also essential
In: Macmillan B. (ed), Remodelling Geography, pp. 204-210. Basil
to consider not only how GIS contributes to geography, but Blackwell, Oxford.
also how geography contributes to GIS research. Curry M., 1998: Digital Places: Living with Geographic Information
Technologies. Routledge, London.
Densham P.J., 1996: Visual Interactive Locational Analysis. In: Long-
ley P.A. and Batty, M. (eds), Spatial Analysis: Modelling in a GIS
Conclusions Environment, pp. 185-206. Geoinformation International, Cambridge.
Dorling D., 1995: A New Social Atlas of Britain. Wiley, Chichester.
Fisher M.M. and Getis A. (eds), 1997: Recent Developments in Spatial
GIS is currently regarded as a tools-science continuum
Analysis: Spatial Statistics, Behavioural Modelling and Computational
(Wright et al, 1997). As Macmillan (1998) points out,Intelligence.
it Springer, New York.
is important to identify GIS technology, science of GIS
Fotheringham A.S. and MacKinnon R.D., 1989: The National Center for
technology and science aided by GIS. The microscope for Geographic Information and Analysis. Environment and Planning A 21:
293-306.
example, is unambiguously artifacts and pieces of technol-
Fotheringham S. and Rogerson P. (eds), 1994: Spatial Analysis and GIS.
ogy. In this case there are two senses in which a microscope
Taylor & Francis, London.
can be thought of as being scientific, that is, it can beFotheringham
used A.S. Brunsdon C. and Charlton M., 2000: Quantitative
Geography: Perspectives on Spatial Data Analysis. Sage Publications,
for scientific purpose, and there is a science associated with
London.
the instrument itself. Similar distinctions can be made in the
Goodchild M.F., 1990: Comment: Just the Facts. Pol. Geogr. Quart. 10,
context of GIS. For instance, the GIS software has a spe-
335-337.

cific design, that develops new methods for handling spatial

Goodchild M.F., 1992: Geographical Information Science. Int. J. Geogr.
Inf. Sci. 6:31-46.
data and spatial analysis. Geographical research is also made
Harvey D., 1989: From models to Marx: Notes on the Project to 'Re-
possible by GIS. Beyond simple observation, new research model' Contemporary Geography. In: Macmillan B. (ed.), Remodelling
topics can start only with the help of new technologies. The
Geography, pp. 211-216. Basil Blackwell, Oxford.
GIS technology can thus rightly be seen as a microscope in
Hearnshaw H.M. and Unwin DJ. (eds), 1994: Visualization in Geographi-
cal Information Systems. Wiley, Chichester.
the field of geography.
Ishikawa Y, 1994: Quantitative Geography on Human Migration. Kokon
GIS has also carried a kind of IT revolution into geogra-
Shoin, Tokyo (in Japanese).
phy. Its influence on geography will be limited if geography
Johnston R.J., (1997): Geography and Geographers: Anglo-American Hu-
can recognize the efficiency of using GIS as a tool that can
man Geography since 1945. Arnold, London. Translated in Japanese, in
Tatsuoka Y, 1997, 1999: Chijin-Shobo, Kyoto.
change a paper map to a digital map. GIS also has a great
Longley P.A. and Batty M. (eds), 1996: Spatial Analysis: Modelling in a
potential to change academic geography to applied geogra-
GIS Environment. Geoinformation International, Cambridge.
phy. The time has finally come when research conducted
Longley P.A., Brooks S.M., Mcdonnell R. and Macmillan B. (eds), 1998:
in academic geography could be applied to attack variousGeocomputation: A Primer. Wiley, Chichester.
Longley P. and Clarke G. (eds), 1995: GIS for Business and Service
contemporary social problems using GIS.
Planning. Geoinformation International, Cambridge.
Macmillan B. (ed.), 1989: Remodelling Geography. Basil Blackwell,
Oxford.
References Macmillan B., 1989: Quantitative Theory Construction in Human Geogra-
phy. In: Macmillan B. (ed.), Remodelling Geography, pp. 89-107. Basil
Blackwell, Oxford.
Bailey T., 1994: A Review of Statistical Spatial Analysis in Geographical
Macmillan B., 1998: Epilogue. In: Longley P.A., Brooks S.M., Mcdonnell
Information Systems. In: Fotheringham S. and Rogerson P. (eds), Spatial
R. and Macmillan B. (eds), Geocomputation: A Primer, pp. 257-264.
Analysis and GIS, pp. 13-44. Taylor & Francis, London.
Wiley, Cambridge.
Batty M., 1976: Urban Modelling: Algorithms, Calibrations, Predictions.
Morikawa H., 1980: Central Place Theory. Taimeido, Tokyo (in Japanese).
Cambridge University Press, London.
Murayama Y, 1990: Regional Analysis. Kokon Shoin, Tokyo (in Japanese).
Berry B.J.L. and Marble D.F. (eds), 1968: Spatial Analysis: A Reader in
Okabe A., 1999: Overview of Academic Journals of GIS and Geographical
Statistical Geography. Prentice-hall, Englewood Cliffs NJ.
Analysis. J. Geogr. 108: 673-677 (in Japanese).

This content downloaded from

134.226.213.129 on Fri, 01 Nov 2024 14:04:19 UTC
All use subject to https://about.jstor.org/terms
180

Openshaw S., 1989: Computer Modelling in Human Geography. In: Taylor P.J. and Johnston R.J., 1995: Geographical Information Systems an
Macmillan B. (ed.), Remodelling Geography, pp. 70-88. Basil Black- Geography. In: Pickles J. (ed.), Ground Truth: The Social Implications o
well, Oxford. Geographic Information Systems, pp. 5 1-67. Guilford Press. New York
Openshaw S., 1991: A View on the GIS Crisis in Geography, or, Using GIS Taylor P.J. and Overton M., 1991: Further Thoughts on Geography and GI
to Put Humpty-Dumpty Back to Again. Environment and Planning A 23: - A Preemptive Strike. Environment and Planning A 23: 1087-1090.
407^09. Wilson A.G., 1972: Theoretical Geography: Some Speculations. Trans
Openshaw S., 1992: Further Thoughts on Geography and GIS Inst. Brit.
- A Geogr. 57: 31^4.
Reply.
Environment and Planning A 24: 463^66. Wilson A.G., 1974: Urban and Regional Models in Geography and
Openshaw S. (ed.), 1995: Census Users' Handbook. Geoinformation
Planning. John Wiley, London.
International, Cambridge. Wilson A.G., 1989: Classics, Modelling and Critical Theory: Human Ge
Openshaw S. and Openshaw C, 1997: Artificial Intelligence inography
Geography.as Structured Pluralism. In: Macmillan B. (ed.), Remodelling
Wiley, Chichester. Geography, pp. 61-69. Basil Blackwell, Oxford.
Openshaw S. and Abrahart R.J. (eds), 2000: GeoComputaion. WrightTaylor &
D.J., Goodchild M.F. and Proctor J.D., 1997: GIS: Tool or Sc
Francis, London. ence? Demystifying the Persistent Ambiguity of GIS as 'Tool' versu
'Science'.
Pickles J. (ed.), 1995: Ground Truth: The Social Implications of Ann. Asso. Am. Geogr. 87: 346-362.
Geographic
Information Systems. Guilford Press, New York. Wrigley N. and Bennett R.J. (eds), 1981: Quantitative Geography: A British
Rogerson P.A. and Fotheringham A.S., 1994: GIS and Spatial Analysis:
View. Routledge & Kegan Paul, London.
YamadaP.M.,
Introduction and Overview. In: Fotheringham S. and Rogerson 1986: The Diffusion Process of the So-called New Geograph
(eds),
Spatial Analysis and GIS, pp. 1-10. Taylor & Francis, London.
in Japan: An Essay on a Contemporary History of Japanese Geography
Sherwood N., 1995: 'Business Geographies' - A US Perspective. In:Reports
Geogr. Lon- Osaka Kyoiku Univ. 24: 1-16 (in Japanese).
gley P. and Clarke G. (eds), GIS for Business and Service
YanoPlanning,
K., 1990: Development of Urban Modelling Research in Britain: A
pp. 250-270. Geoinformation International, London. Citation Analysis. Human Geogr. 42: 1 18-145 (in Japanese with English
Tanabe K. (ed.), 1982: Urban System in Japan: A Geographical
abstract). Study.
Kokon Shoin, Tokyo (in Japanese). Yano K., 1999: World of Geographical Information Systems: What's GIS1
Taylor P.J., 1990: Editorial Comment: GKS. Pol. Geogr. Quart. 9: 211-212.
Newton Press, Tokyo (in Japanese).

This content downloaded from

134.226.213.129 on Fri, 01 Nov 2024 14:04:19 UTC
All use subject to https://about.jstor.org/terms