War can be defined in many ways, generally any organized conflict with a political basis is considered war. The subtypes of war depend on the dynamics between parties and on the nature of the conflict. Many aspects of defining a war can be debated, like how many deaths make it a war, if cyber-attacks count, and when a war starts and ends like with the Cold War. Organizations such as the Correlates of War (COW) or the Peace Research Institute in Oslo (PRIO) provide definitions for war, and its subtypes, with certain prerequisites that it must fulfil. The nature of a war will impact how it is defined, meaning we look back on certain conflicts differently.
Core Definitions of War
Building on the general understanding that war involves organized, politically motivated armed conflict, scholars and data projects have sought more precise criteria. A widely used threshold comes from the Correlates of War project, which requires sustained combat between organized forces resulting in at least 1,000 battle-related deaths in a calendar year (Sarkees and Wayman, 2010). This numerical benchmark helps distinguish war from lesser forms of violence, such as riots or skirmishes. The Peace Research Institute Oslo applies a lower threshold of 25 battle-related deaths per year for an armed conflict, allowing researchers to capture a broader range of episodes (Gleditsch et al., 2002). These differences illustrate how even basic definitions shape what counts as war in the first place.
Debates persist around the boundaries. For instance, does a cyber operation that disrupts critical infrastructure but causes no direct fatalities qualify? Traditional measures centred on battle deaths struggle with such cases, potentially leaving hybrid or non-kinetic actions outside conventional typologies. Similarly, the Cold War’s start and end dates remain contested because overt combat thresholds were never crossed, yet the period featured proxy wars and ideological confrontation on a global scale.
Typologies and Subtypes of War
Once a conflict meets the definitional criteria, typologies further classify it according to the parties involved and the stakes at issue. A basic distinction separates interstate wars, fought between two or more recognized states, from intrastate or civil wars, occurring within a single state’s borders. The Correlates of War dataset records both categories separately, noting that civil wars have become statistically more common since 1945 (Sarkees and Wayman, 2010). Additional subtypes include extra-state wars involving a state and a non-state actor outside its territory, and wars of independence, often treated as a distinct category within civil conflict data (Gleditsch et al., 2002).
These categories are not merely descriptive; they carry analytical consequences. Classifying a conflict as interstate tends to direct attention toward balance-of-power dynamics and alliance behaviour, whereas intrastate conflicts prompt examination of ethnic grievances, resource distribution and state weakness. Consequently, the subtype assigned influences both the questions scholars ask and the policy tools considered appropriate.
Effects on the Conduct of War
Definitions and typologies also shape operational realities. When governments recognise a situation as an interstate war, they typically mobilise regular armed forces, seek formal declarations or parliamentary approval, and apply the full body of international humanitarian law. In contrast, conflicts labelled as counter-insurgency or counter-terrorism campaigns may involve special forces, drone strikes and partnerships with local militias, measures that sidestep some conventional legal and political constraints. The decision to treat hostilities in Afghanistan after 2001 as a non-international armed conflict, for example, affected rules of engagement and detention policy.
Furthermore, threshold requirements can encourage actors to calibrate violence. A state wishing to avoid triggering the 1,000-death benchmark may limit the intensity or duration of operations, while rebels might seek to escalate casualties precisely to gain recognition as a warring party. Thus, the act of definition feeds back into strategic behaviour on the battlefield.
Effects on the Study of War
In academic and policy research, definitional choices determine the composition of datasets and, by extension, the patterns that emerge. Because COW’s stricter fatality threshold excludes many low-intensity conflicts, statistical analyses based on its data tend to under-represent conflicts in sub-Saharan Africa or Latin America after the Cold War. PRIO’s lower threshold produces a larger sample and highlights the prevalence of civil wars, altering conclusions about the so-called “democratic peace” or the impact of peacekeeping operations (Gleditsch et al., 2002). Researchers must therefore remain aware that findings are partly artefacts of the coding rules employed.
These coding differences also affect forecasting models used by governments and international organisations. A model trained on COW data may overlook simmering insurgencies that later expand, leading to surprise when violence suddenly crosses the higher threshold. Policy prescriptions derived from such models can therefore rest on incomplete pictures of contemporary violence.
Conclusion
In sum, war and its subtypes are defined through a combination of political intent, organised violence and quantitative thresholds, yet these criteria remain subject to debate and revision. Different typologies steer both the tactics states adopt and the research designs scholars employ. Because definitions are never neutral, analysts and practitioners must continually interrogate the categories they use, recognising that how a war is named helps determine how it is fought and understood.
References
- Gleditsch, N.P., Wallensteen, P., Eriksson, M., Sollenberg, M. and Strand, H. (2002) Armed Conflict 1946–2001: A New Dataset. Journal of Peace Research, 39(5), pp. 615-637.
- Sarkees, M.R. and Wayman, F. (2010) Resort to War: 1816–2007. Washington, DC: CQ Press.
