In the twentieth century, threats to security were measured in terms of the military capabilities of potential enemies. Publications like the International Institute for Strategic Studies (IISS) Military Balance would provide lists of capabilities of NATO, the Warsaw Pact and non-aligned nations including trends in personnel, military equipment and budgets. The aim of this component of the programme is to construct a date base of alternative indicators of insecurity better suited to the contemporary period.

There have, of course, been many efforts to develop alternative indicators of insecurity or global risk. Because this is a new field, the effort is fraught with definitional and empirical difficulties. There are, for example, new attempts to measure the incidence of violent conflict. The Uppsala Conflict Data Program (UCDP) provides numbers that are used both in the Human Security Report and the annual Stockholm International Peace Research (SIPRI) Yearbook (SIPRI various years, UCDP various years). The UCDP data was initially based on a twentieth century notion of conflict. For a conflict to be included in the data, there had to be at least one thousand battle-related deaths and a state had to be involved. This is highly problematic since battles are rare in twenty-first century conflicts and conflicts usually involve non-state actors or a mixture of state and non-state actors. More recently, UCDP have updated their data to include conflicts involving non-state actors and to include what they call one-sided violence, i.e. violence against civilians. In addition, recent versions of the Human Security Report also include some measure of human rights violations, in particular, the use of child soldiers. But they still mainly count battle deaths as casualties and provide insufficient information about direct civilian casualties (from violent attacks) and indirect civilian casualties (from lack of access to food and medicine, the spread of disease in conflict zones and the loss of livelihoods). Moreover, they depend primarily on media reports, which tend to underreport deaths and are rather patchy.

International agencies, particularly UNDP, have attempted to measure a broader concept of security, namely a holistic interrelated set of factors that affect the wellbeing of human beings. Thus, for example, the Arab Human Development Report for 2009 addresses the challenges to human security in the Arab region (see Report and additions at It includes qualitative and quantitative data on seven dimensions of security. These are: people and their environment; the state and its insecure people; the vulnerability of those lost from sight (women, children, displaced persons, etc); volatile growth, high unemployment and persisting poverty; hunger, malnutrition and food insecurity; and health security challenges. The report provides an invaluable documentation of human development failings in the Middle East that would undoubtedly need to be studied by those identifying insecurity in the Middle East. But by being so broad, there is perhaps an inadequate emphasis on physical threats to security that are most pressing for people who experience them; these are not only deaths from violent conflict as in Gaza – they have to do with the breakdown of or disruptions in law and order. Moreover, socio-economic indicators are more readily available from such sources as UNDP, the World Bank, or the United Nations Millennium Development Goals.

There is a set of indicators that need to be developed that address this physical aspect of insecurity. As well as violent conflict, these include human rights violations, violent crime, domestic violence, refugees and displaced persons and casualties from natural disasters (floods, hurricanes, earthquakes, etc.). A number of places, both governmental and non-governmental, collect data on human rights violations (Human Rights Watch, Freedom House, Amnesty International, the US State Department, the UK Foreign and Commonwealth Office) but the results tend be anecdotal and difficult to compare. The major database for crime statistics is the UN Office of Drugs and Crime (see It collects data via questionnaires addressed to governments. However, as typical with these kinds of data, coverage is weakest in those countries experiencing greatest insecurity. Other possible sources include statistics collected and supplied by national governments – these vary greatly in terms of availability and coverage.

Obtaining comparable data on domestic violence is extremely difficult, for a number of reasons. Definitions vary between sources, and so therefore do measurements. Currently, there is no central or major database for this indicator. Case studies exist for individual countries, and it is possible to draw information from reports by Amnesty International, Oxfam, local women’s groups, etc., but there is no repository as such for ‘hard figures’. Official statistics are scarce, especially in developing countries. In many regions, any information on the incidence of domestic violence exists only in ‘qualitative’ form, reported from interviews, etc.

Data on refugees and internally displaced persons is more comprehensive. The United Nations High Commission for Refugees provides detailed statistics on refugees and internally displaced persons as does the United States Refugee Committee. The Internally Displaced Monitoring Centre in Geneva also traces internally displaced people (see According to UNHCR’s 2009 Global Trends Report, some 42 million are uprooted worldwide; 16 million are refugees and some 26 million are internally displaced people.

Finally, the main source for victims of natural or man-made disasters is the international disaster data base at the University of Louvain in Belgium (see Guha-Sapir et al. 2004). A general problem with this data collection, like in any database that builds on various sources, is the dependency on the individual data suppliers and their accuracy – even more, since the original data is not specifically collected for statistical purposes. In addition, data for a number of countries are not available in the database at all.

The indicators component of the programme will aim to pull together this miscellaneous collection of data, and to develop methods that can potentially fill lacunae and improve accuracy. It is extraordinary that, whereas information about military casualties, where regular forces are involved, is acccurate and detailed, in most conflicts, even the order of magnitude of civilian casualties is disputed. Moreover, statistics that could assist planners in identifying areas of actual or impending insecurity are, at present, quite inadequate.

Two approaches have some promise. Firstly, in asessing areas of insecurity world wide, data on population displacement (refugees and internally dislaced persons) could be used as proxy indicators. People are displaced deliberately as in wars and genocides; they flee danger; they look for work and food where livelihoods have been destroyed or they flee from natural and man-made disasters. And the data sets are relatively good. Secondly, in local sitations, the factors that contribute to human insecurity vary. The distribution of insecurity is not uniform even within a particular area, neither in terms of levels of threats, nor in terms of types of threats. One method that has been proposed to deal with this variation is to use Geographical Information Systems (GIS) software to map the specific combination of factors that are relevant in a particular area, their relevance being determined by consultation with actors local to the region (Owen 2006).

The programme will test out these different approaches and also collect its own primary data, both from case studies undertaken in other components and through the use of the Internet as a research tool. How this is done will be described in the methodology section, but the main utility of this new approach is that it greatly expands the geographical and analytical scope of data on insecurity and it allows those who experience insecurity to shape the categorisation of indicators.