It is useful to start by recognising the following situation. GIS is here to stay and really it matters little whether or not some geographers and social scientists do not like it much. Nevertheless, it is worrying that so many GIS applications are being performed in a socially naive manner by users who are largely oblivious to the potentially broader implications of what their technology is being or maybe used for. It is also worrying that in attempting to improve this neglect that so much ill- informed rubbish is being written about GIS, presumably by geographers and others who simply do not understand what it can and cannot reasonably be expected to do, or else are motivated by other concerns that result in them viewing GIS from perspectives that continually seek to question its legitimacy. Perhaps the principal problem with these critiques is that they contain an almost random mixture of trivia with a few very significant concerns but without much or any distinction in importance between them. It is almost as if the act of being critical will somehow, magically doom, GIS to the same oblivion that quantitative geography was apparently dispatched to a quarter of a century earlier.
It is useful to recognise that in excess of 90% of all GIS
applications, maybe even 99.9%, are of no significant consequence
to people and society. They involve applications that are
concerned with the management of the physical infrastructure;
such as those involving drains, wires, pipes, parcels of land,
and roads. The vast majority of all GIS applications involve
little more than a digital replacement for various large scale
paper map making, recording, and handling industries. The
problems with GIS that are of broader relevance to society lie
elsewhere. In fact it might be useful to restructure the current
critiques to focus on the role of GIS in three different domains
which are enumerated as follows:
(1) the wider role of GIS in ongoing IT developments affecting
the whole of society with the emergence of IT States;
(2) the nature of GIS as a database technology being used to
represent information about people; and
(3) the use of GIS in spatial decision support applications that
impact on people.
This threefold categorisation offers a potentially useful
rationalisation of the various debates about GIS and society.
GIS adds an important and significant new dimension to many IT
developments. However, what is the vital contribution that GIS is
making to the continued development and eventual appearance of
the IT State? Certainly, in most countries of the world, there
are major changes underway in IT. When viewed in isolation these
changes often appear to be mundane and fairly harmless
technologies; for example, fibre optic cabling of homes, the
gradual integration of multimedia technologies, and the falling
cost of hardware permitting its use in an increasing number of
domestic products. However, when these separate developments are
viewed or linked together then a very different picture emerges
with a unique potential for both good and bad on a scale and at a
level never before encountered. Openshaw (1992) noted that
"Too many people are ignorant of what is now possible and
have no concept of any need to develop control systems designed
to preclude misuse. The few exceptions form an information elite
who could well become the dictators of the information age...
Modern IT can indeed be used to strengthen and improve
representational democracy but it also provides the basis for the
most powerful and effective forms of 'people control' of the sort
never before seen in history" (p 104). He outlined a five
stage model of IT development in which GIS occurs in every stage:
Stage 1: gradual computerisation of everything over a 50 year
period including the cartographic industry and the ability to add
locational references to personal data;
Stage 2: distributed computing and open systems enable the
linkage of historically separate systems whilst in GIS the
development of data standards allowing ease of data transfer;
Stage 3: expansion down to the domestic level and covering all
aspects of life as databases become broader in coverage so the
value added by geographic file linkage mechanisms becomes
significant;
Stage 4: construction of systems design to provide an improvement
in well being and public good and it can be argued that automated
exploratory spatial analysis tools have a role to play in
monitoring GIS databases for patterns, relationships, and the
detection of anomalies which might be "useful";
Stage 5: potentially malevolent applications that could infringe
civil liberties and human rights involving people surveillance
and ultimately people control systems in which the ability to
track movements in space is one of the key component
technologies; and
Stage 6: an ultimate end state in which regulated, integrated,
people management and control systems extend into all areas of
existence and become a modern necessity in order to survive,
perhaps following one or other major catastrophes.
The stages overlap and maybe Stage 3 has already been reached with aspects of Stages 4 and 5 becoming evident. The driving force is that of a continued and still increasing process of technological change and development, combined with the desire of governments, businesses, and people to survive. In IT terms we are still in a state of transition to an information state and GIS is one of the key component technologies but it is not the only or even the prime one. Yet if GIS did not already exist then it would, sooner or later, have been invented because it is so obvious and constitutes a key infrastructure resource. It is, therefore, unavoidable and inevitable that sooner or later GIS will become an important part of the set of modern management tools available to States and big business. In an increasingly political unstable world with problems of ever increasing complexity, it is inevitably that countries will use whatever means are available to ensure their survival. The problems for society concern what to do about these IT developments (not all of which can be perceived), how it might be controlled, and how to erect barriers that will stop an information elite gaining power for themselves. At the same time how do we balance these concerns with the potential harm that neglect of the technology might engender; for instance a State that fails to gain maximum benefits from IT might well be considered to be committing another type of crime. Openshaw (1993, 1994) identifies various types of GIS and spatial analysis crime and argues that the deliberate non-use of available technologies is a crime of some sort but so too is its abuse and naive usage. Surveillance may well be considered an infringement of public freedoms and liberties but it might also save lives, reduce crime, and result in more efficient government. The really fundamental question is how to get the balance right, how to manage the systems to minimise potential abuse, and how to future proof the barriers so that they cannot easily be subverted.
GIS technologies constitute only a small part of a much bigger
picture that is being driven by a virtually unstoppable process
of technological change. An equilibrium state has not yet been
reached. The problem is that end user appreciation of what is now
possible is lagging far behind what is now feasible, and what is
now considered feasible is itself far behind what will soon
become possible. At present, anarchy rules and history suggests
that at times like these dictators often emerge!
At a more detailed technical level, concerns exist about the nature of the representations of people provided by current GIS data models. Some critics emphasise various weak aspects of GIS that most GIS users would consider to be self obvious; for example, that GIS is map based, that maps are not necessarily accurate nor objective representations of reality, maps reflect the value systems of the map makers and have a historical context, States for various usually innocent reasons often distort what are shown on maps, the current GIS data models do not necessarily provide a good representation of people rather than property and inanimate infrastructure, the technology is not evenly distributed, it is being used by the military and States to "target" various people and areas; and there are no real controls on its use. There is a danger that the users of GIS will be misled into thinking it is a totally objective technology that provides a value free, scientifically correct and rigorous view of the world; when there are all manner of possible theoretical and methodological problems that may sometimes matter. To some extent the problem is knowing how to separate the important issues from the irrelevant distractions.
There are other questions that can be linked to these concerns. In particular "Can there be a socio-economic GIS?" or "What needs to be done to existing GISs to improve their capabilities in handling data about people?" or "Is the concept of socio-economic GIS so ridden with contradictions that such a system could never be built?" or "In the IT age what sorts of socio- economic database system would be regarded as necessary if the problems with GIS and conventional technologies are to be avoided or reduced?" or "Can there be a qualitative version of GIS?". Its not too late to invent a new technology if there are good reasons for doing so. There are major developments underway in soft computing technologies that may be very relevant in developing new types of GIS orientated towards social rather than map spaces.
Certainly in GIS the conventional data cube approach is still alive and well and in one form or another underpins most of current database technologies. Computerisation of details about people ultimately reduces to measuring the presence or absence or amounts of predefined variables that can be data captured ideally without too much difficulty by automated means. The emphasis is still firmly on accessible and thus partial quantitative information with all the well known problems but the appearance of soft computing technologies able to capture, store, and analyse video and audio data expressed in a digital form should not be underestimated. In some ways it is fairly trivial to add a multimedia dimension to the spatial information database. Equally it is possible to imagine GIS technologies being applied to capture and represent much more of the invisible and less tangible domains of the cyberspace with or without a geo component to it.
It is an appropriate time to consider how to specify (and
perhaps build) alternative systems rather than just moan
endlessly about the problems associated with those that exist.
There is a political economy and social context to many GIS applications that the non-sensitised end-user will probably never have noticed. Most spatial decision support applications of GIS involve attempting to change or tampering with the real-world. There is nearly always a set of benefits that are sought and a cost to be born in attempting to achieve them. The challenge for the spatial scientist has traditionally been how to develop improved technologies that will yield even 'better' results with the definition of 'betternness' being left to others. The challenge for the social scientist is to provide a means of using GIS, of discovering ways of debating with it in order to address at least some of the broader concerns associated with its use. GIS is often viewed as favouring the establishment, the powerful, and the rich whilst harming in various unmeasured ways the poor, those who lack access to the technology, and have no spatial information of their own. This view is of course highly simplified and there are counter examples whereby minority groups have used GIS as tools to criticise governments and big business; for example, Openshaw's (1986) critique of nuclear power siting and Openshaw et al's (1989) critique of radwaste site selection.
There is a fundamental inherent contradiction in examining the impact of spatial decision support systems. Is it better not to use an SDSS than it is to use one? For example, suppose two banks merge and 20% of the branch network is to close. Do you: (a) use a GIS with an embedded spatial network optimisation procedure that seeks to maximise network profitability by determining the 'best' branches to close? or (b) do you select branches to close so that access by social disadvantaged groups (who may contribute little to bank profits) are preserved; or (c) muddle through doing neither using pre-GIS technology. The dilemma is that whilst (a) may well allow the bank to prosper after downsizing and secure the continued employment of 80% of the staff, both (b) and (c) may result in a continuing crisis with further branch contractions and closures. Its not easy but if these broader social implications are to become important then some practical means needs to be developed to allow them to be articulated in a form the geoprocessing technologies can handle.
Likewise, governments presumably have a duty in managing society to use whatever information resources exist; for example, in resource allocation, detection of crime, and prevention of terrorism. The geographic linkage of personal data is clearly an important activity that has the potential for both good and harm. It is perhaps worrying that at present the very state agencies most likely to be interested and actively involved in violating the confidentiality of personal information are (a) exempt from whatever Data Protection legislation exists, and (b) they will always be exempt on state security grounds. By comparison what the commercial marketing research sector attempts to do is pathetic in scale, grossly exaggerated in terms of its surveillance capabilities, and usually of a poor quality. Yet its quite fashionable to criticise the commercial sector whilst totally ignoring the security services. It is also fashionable for government departments, fearful of bad press publicity, to deliberately not-use key information resources at their disposal. In the UK there are many databases for which an undisputed public good imperative for analysis and use exists but with little attempt to either analyse or use the data; for example, monitoring of spatially reference mortality and morbidity data for anomalous patterns. Public expenditure is still being distributed using data which is at best 5 years out of date and can easily reach 12 years out of date.
There are clearly some important methodological problems that also need to be resolved. For example, what is a safe level of spatial aggregation for presenting data about persons? In the UK the confidentiality risks present by different levels of geography have yet to be quantified. Data are released in a form that may well be too safe and in the process become so damaged as to impair many of the subsequent applications of it. Methods are needed to optimise the statistical disclosure risks.
Another area to consider is how best to use GIS as an aid to democracy. How should GIS be used as a public debating device? If its use is one-sided then it is clearly unfair, so what procedures and what mechanisms are needed so that its use can be made fairer and available to all? Openshaw and Carver (1994) argued that the key here might be the development of decision explaining systems, and that maybe there should be a requirement on the users of GIS to provide systems that allow others to explore the alternatives, to interrogate the databases, and to test out their own scenarios. Of course this may amount to no more than a public relations exercise but the new technologies that allow decision makers to optimise their decisions should also be available to those who wish to debate and influence them. Maybe this will be easier in a public policy context rather than in commerce, except that even within organisations there is a need for informed debate and dissension. Perhaps the key aspect is to distribute the information and the GIS tools rather than concentrate it in the hands of very select groups.
The challenge is to identify mechanisms that will allow the
end-users of the technology to at least understand the areas of
concern in those applications where they matter most as distinct
from those where they may not matter at all. If this is to work
then the language used to communicate these concerns has to be
understandable to the end-users and developers of GIS. There is
also a need for illustrations to show that they matter, rather
than being of purely academic interest.
The success and use of GIS on a global scale has proceeded
with a minimal amount of debate about the broader impacts on
society. This is quite understandable because of the historic
origins of GIS in land and property information systems. However,
it is clearly important that as the technology starts to be used
and more directly affect the environment and lives of people that
there is a much better understanding of how to use it in an
intelligent, sensible, and sensitive way. Currently, in common
with the rest of IT, GIS is uncontrolled and unregulated other
than indirectly by the access of spatial information and by the
inherent limitations of the current versions of the technology.
GIS is not going to go away so the only real option is to develop
a means of using it that retains a sense of balance and is fair
to those who may be affected by it. It is not helpful to merely
argue against GIS or any form of geography that uses it. The real
concerns for the future are much more important than parochial
debate about this or that species of human geography or social
anthropology. There is not much time left and it is very
important to consider ways of defining ground rules, good
practice, and protocols that will provide at least a degree of
regulation to a GIS industry that is currently largely out of
control. Of course it may not matter, but if it does not matter
at one level of IT then who is to know what the follow-on
implications may be at another level?
Openshaw, S., 1986, Nuclear Power Siting and Safety. Routledge and Kegan, London
Openshaw, S., 1992, 'Some aspects of Information Technology and the future', Proceedings of AURISA 92, p104-113 Australasian Urban and Regional Information Systems Association
Openshaw, S., 1993, 'GIS crime and criminality', Environment and Planning A 25, 451-458
Openshaw, S., 1994, 'GIS crime and spatial analysis', in Proceedings of GIS and Public Policy Conference, Ulster Business School p22-35
Openshaw, S., Carver, S., Fernie, J.,1989, Britain's Nuclear
Waste: Siting and Safety. Belhaven, London