How convincing is the argument that democracies don’t fight each other?

There is a liberal theory in international relations called the democratic peace proposition, also known as the democratic peace theory. Within this theory, there is a monadic and dyadic hypothesis, stating that either democratic states are less violent in their foreign relations in general, or that they are specifically less violent in their relations with other democracies. The proposition itself states that democracies rarely, if ever, go to war with one another. The support for the theory that democratic states do not go to war with other democracies is, according to Bruce Russett, “one of the strongest non-trivial or non-tautological generalisations that can be made about international relations” (Chan, 1997, p.60). This is not a new or unheard of conclusion, with Bill Clinton noting that “Democracies rarely wage war on one another”, and former NSA advisor Anthony Lake observing that democratic values and international peace mutually reinforce one another (Chan, 1997, p.59). This hasn’t always been the case, however. Renowned American political scientist Quincy Wright stated quite uncontroversially (at the time of 1942) that there were warlike and peaceful tendencies in both democracies and autocracies, and statistics wouldn’t validate a democratic peace hypothesis (Ray, 1995, p.30). But in modernity, this belief of the sanctity of democratic peace has become a linchpin of American thought, and that promoting demcratic institutions across the world is to make a more peaceful international society (Kissinger, 1994, p.33). In this paper I will demonstrate that, if ‘fighting’ means war, the argument that democracies do not fight each other is a very convincing one. I will also attempt to analyse why this is the case as well as other contributing factors, and what it means for the realist school of thought. I will be primarily using the term ‘fight’ to mean inter-state armed conflict, but other violent or subversive actions will be looked at. Additionally, I will endeavour to evaluate what we consider as a ‘democracy’, and other interpretations of such.

The idea of a democratic state is a continually developing concept. What was considered a democracy a century ago may not be considered one now (Ray, 1998, p.31). The breakaway Confederate States of America was arguably a democracy, but one in which less than 50% of the population could vote. Even modern definitions of democracy are debated, with Dr Ido Oren stating that democratic ideals are instead the products of America’s past foreign relations (Chan, 1997, p.65). That, however, is beyond the scope of this paper. When agreeing that democracies have not gone to war with each other, it is an agreement that states which have achieved a certain level of democracy have not done so (Ray, 1998, pp.32-33). While Jack Levy describes the absence of war among democratic nations as being the closest thing to an empirical law in international relations, this was not always mainstream scholarly thought. Sociologist Dean Babst wrote what was at the time an obscure article in the journal Industrial Research noting that between the period of 1789 to 1941 no democracies had gone to war with each other (Chan, 1997, p.60). The initial consensus in International Relations remained that there was no particular correlation between the regime type of a state and how war-prone they were held until challenged by Rudolph Rummel. Rummel focused on what he called libertarian states, or states emphasising civil liberties alongside a competitive election of leaders, which is a close enough comparison to liberal democracies. He found that these libertarian states were less involved in war than their autocratic counterparts, which is disputed, and they were never involved in war with another democratic state, which is more universally recognised (Chan, 1997, p.63).   

Proponents of the democratic peace theory and the theory that democracies do not go to war with each other hark back to Immanuel Kant’s Perpetual Peace. Kant theorised that perpetual peace overcoming international anarchy was possible through republics respecting the will of the people. Public opinion would be a force against belligerence as the citizenry would be reluctant to commit to “such a poor game, decreeing for themselves all the calamities of war” (Kant, 1795). Commerce and shared values and common institutions would be the two other factors that would ensure a perpetual peace. Rummel favoured a monadic hypothesis finding a negative relationship between democracy and violence as a whole, not just between democracies themselves. Theorists such as Kenneth Boit and Jon Wilkenfield support Rummel’s view that democracies reduce the odds of conflict in general, not just in conflict with non-democracies (Chan, 1997, p.63). This monadic view is not accepted by most theorists who prefer a dyadic perspective of a democratic peace- that democratic states create a separate and joint zone of peace amongst themselves that does not extend to autocracies. David Rousseau found significant evidence in support of a democratic peace but not in a peace encompassing both democratic states and all of their foreign relations- just the democratic ones (Chan, 1997, pp.63-64). This does not necessarily disprove Kant’s theory. Democratic states will still have to participate in power politics in the international system as they operate in a non-Kantian system alongside major nondemocratic powers. The studies advocating a democratic peace are not done in a vacuum either, such as the study by Maoz and Abdolali that found no inter-state wars between democracies, and less between democracies and other states than expected. This was adjusted for geography, economics, and alliances among other factors. The results still show that out of all of the possible variables, only the regime type and political instability factor into a state’s war-proneness (Maoz and Abdolali, 1989, pp.3-35).

The argument that democratic states do not fight each other, and the democratic peace theory in general, is not without its detractors, both to its premise and its implications. One of the main issues is that until recently democracies have been a minority, and a minority in an international system with far less nation-states. In 1840 there were 32 sovereign states, and today there are approximately 196, depending on the recognition of partially recognised states and nations of disputed sovereignty (Chan, 1997, p.74). Of these, the 21st century is the first time in history in which the majority of them are democracies with 57% being assessed as democratic, albeit many of them having shaky democratic foundations. The number of outright autocracies in the world is only 13% (Desilver, 2019, online). Skeptics question whether data analysis and pattern identification of such a limited data set is viable. The probability of war at any one time is low, but even with a cumulative effect the time period in which democracies have been a significant proportion of the international system is miniscule.  Even if democratic states had the same propensity for war-proneness as autocratic states, with the time constraints and number of possible and reasonable democratic pairings for potential war, the expectation for any state to engage in war would be low (Chan, 1997, p.72). We are also limiting the word ‘fight’ in this context to warfare. 

Studies done by Lewis Richardson and Quincy Wright in 1960 and 1942 respectively found that democracies were not less prone when the definition of conflict was expanded beyond war to escalation, skirmishes, or military interventions (Richardson, 1996). As seen with the covert actions of the West- notably the USA- during the Cold War, if ‘fight’ were to also include subterfuge, espionage, and subversion, democracies would look less peaceful. This was not limited to actions just against autocratic states either. The Guatemalan Revolution in 1944 resulted in the first free elections in the country and a successful and peaceful transition of power between two democratically elected presidents (Gleijeses, 1991, p.3). The Eisenhower Administration saw Guatemala as siding with the Soviets and began a campaign to discredit the Guatemalan government politically before launching a coup using CIA armed and backed right-wing guerillas to overthrow the elected government (Immerman, 1982, p.110). Although not a formal war by any means, one could certainly expand ‘fight’ to include such a campaign. The definition of war and what is counted as a war is problematic in and of itself. The threshold for a state to be counted as a participant in a war is low when contrasted with the threshold for war itself- considered as having 1000 personnel deployed in the conflict or 100 KIA by many such as Arvid Raknerud and Havard Hegre. Equally, the 100-hour ‘Football War’ between Guatemala and Honduras would be counted as one incident of war, as would Vietnam. These factors result in data analysis of wars democracies have been involved in not necessarily representing the true magnitude and reality of what occurred.

Although I have decided to limit the definition of ‘fight’ in this instance to interstate wars, there is an argument to be made against the democratic peace theory that it should include civil and extra-systematic wars. After all, the realist scholar Christopher Layne opined “if democratic norms and culture fail to prevent the outbreak of civil war within democracies, what reason is there to believe that they will prevent the outcome of interstate wars between democracies?” (Layne, 1994, p.41). Layne goes on to challenge perceptions that public opinion will always be an inhibitor to democratic states going to war. In 1891, the McKinley Administration was reluctant to engage in war with Spain, but was encouraged to do so by public opinion. In 1914 France and Britain, public opinion was enthusiastic about a perhaps inevitable war with Germany (Becker, 2015, online). Regardless, if we limit ‘fight’ to war, are there any examples of a clear cut case of an interstate war between democracies? Rummel did not count marginal or ambiguous cases against the democratic peace theory, but stated that “one clear case of violence or war unqualified by very unusual or mitigating circumstances falsifies the proposition [of democratic peace]” (Chan, 1997, p.70).

For one possible instance there was the Cenepa War lasting around a month with less than 100 official casualties between Peru and Ecuador. Ecuador was a solid democracy, however Peru was led by Alberto Fujimori who took control via a coup and won a second term with accusations of voting irregularities (Nohlen, 2005, p.454). A better example on the South American continent may be the Paquisha War fought between Ecuador and Peru, two young democracies, in 1981. Do these states pass the democratic threshold? Perhaps not, as neither nation had democratic control over their militaries (Weart, 1998, p.70). The Turkish Invasion of Cyprus was in response to a Cypriot coup d’état (Mansfield and Snyder, 2007, pp.223-225), the First Kashmir War was between two supposedly democratic British Dominions although in reality the Pakistani military was not under civilian control (Paul, 2005, pp.47-48), and Lebanon in the Six Day War was a very disputed democracy (Systemic Peace, 2020, online). The Continuation War involved democratic Finland and Britain and the US, although was principally against the Soviets and according to historian Jonathan Clements was mainly for appearances sake resulting in few casualties (Clements, 2012, pp.208-210). While these could fit a definition of a ‘fight’, none of these examples are clear-cut cases. There has not, to date, been such a clear-cut case between two states with developed liberal democracies without mitigating circumstances (Chan, 1997, p.60), despite opponents of the democratic peace proposition pointing out potential irregularities and flaws in both the data analytics and theory. But if democracies do not fight each other, why is this the case? Additionally, there is conflicting evidence about whether governments with more constraints, inherent in many liberal democracies, are the least war-prone. Clifton Morgan and Sally Campbell in 1991 found that governments facing the most severe constraints on decision-making were the least war-prone (Campbell and Morgan, 1991, p.190). Conversely, Edward Mansfield and Jack Snyder found that too much constraint lead to hypernationalism, exemplified in democratising states such as Weimar Germany (Mansfield and Snyder, 1995, pp.12-13).

It is theorised that democracies will only extend their domestic norms of tolerance and compromise to other states they perceive to be a democracy (Starr, 92, p.208). Studies have shown that political ideology (as well as religion) distinguishes allies from adversaries in the eyes of the population, and that public opinion is more inclined towards the use of force against an autocracy than a democracy (Ray, 1995, p.93). Bueno de Mesquita and David Lalman theorised that there is a dichotomous category in regards to whether two democracies will go to war, and that is whether both states perceive the other to be a fellow democracy, or ‘democratic dove’ (Mesquita et al, 1986, p.158). Leaders perceive others to be either a fellow democratic dove, an ambiguous case, a nondove, or a ‘hawk’. When there is mutual recognition of dovishness, leaders prefer to negotiate than fight. They are, however, likely to err on the side of caution and place a dove in the nondove category rather than risk the other way around (Mesquita et al, 1986, pp.160-165). John Owen noted that “a liberal democracy will only avoid war with a state that it believes to be liberal” (Owen, 1994, p.102) 

There are also both structural and normative explanations for why democracies avoid war with each other. For a structural explanation, there are structural restraints and shared cultures in democracies. The institutions of democracies provide significant numbers of checks and balances as well as limiting the autonomy of leaders  (Choi, 2003, pp.143-145). In unrestrained autocracies, there are far less checks to a leader’s power, and if set on war, there is little that can be done to prevent it. This, however, seems an imperfect explanation of why democracies do not go to war as there are various other nondemocratic forms of state which through institutions and other arrangements limit the power of the leader. If the institutions that restrain leaders and executives prevent democracies from going to war, they ought to prevent all states with similar constraints which is not the case. Moreover, it is often the case that these democratic structures- cabinets, legislatures, and publics, have been more belligerent than the government heads they are supposed to constrain (Mintz and Geva, 1993, p.495). The normative explanation is likely a better answer. Democracies do not intrinsically treat each other ‘better’ than other states, but they are less likely to go to war with them (Weart, 1994, p.485). This is theorised to be because of the domestic norms that exist in developed democracy that allow for it to work effectively such as the willingness to compromise, the contestation of elections, power sharing, and tolerance towards differing viewpoints. If both states recognise the other as a fellow democratic dove, these domestic norms that are conducive to a democratic culture are expanded to cover inter-state relations (Mintz and Geva, 1993, p.485). As they are conducive to nonviolent contestation of power, they are conducive to a similar nonviolence contestation between fellow democratic states. Within normative theory, some recognise the current international system as a unique opportunity to reconstruct international norms and values that underpin the international system to better reflect the interactions between democracies which now make up a majority of the world’s systems (Russett, 1993, pp.279-280). It is also important to note that not all norms are positive, and a widening chasm between social and liberal democracies may see greater tensions in the future (Widmaier, 2005, p.435).

The democratic peace proposition is also an affront to the realist view of international relations, and traditional realist thought cannot coexist with a democratic peace- “realism has no place for an expectation that democracies will not fight each other” (Russett, 1993, p.24). Realism proposes that each state is first and foremost a selfish entity that will act in its own interests for power or safety in an anarchic self-help system. The democratic peace proposition, or even the basic idea that democracies do not or very rarely go to war with each other has democracies going against their own interests in the interest of peace. The regime type of a state, and whether it is a democracy or not, ought to have little bearing or constraints on states acting in what they perceive to be the most advantageous way. Mutual democracy, the democratic peace theory proposes, constrains states sovereignty in a way that is incompatible with realism. Realist theories to counteract this include that the system of alliances such as NATO prevents war between democracies, not their democratic nature, but this is an incomplete explanation in a world full of unaligned democracies.

There is evidence, however, that democracy may be a poison chalice- at least initially. While evidence suggests that democracies do not go to war with each other, it also indicates that democratising states (states that are undergoing a transition towards democracy) are more likely than either democracies or nondemocracies to go to war (Russett, 1993, p.636). Democratisation could consolidate a new democratic regime, but also forms a high likelihood of triggering hypernationalism as seen in Wilhelmine Germany and with Serbia’s Slobodan Milosevic (Zakaria, 1997, p.35). A state is more likely to succeed in forming a stable democracy in place of a nondemocratic regime when there is a particular historical sequence- the emergence of a national identity alongside the institutionalisation of the central government, and followed by mass electoral and political participation (Mansfield and Snyder, 2006, pp.36-38). This is an ideal scenario, but is an unlikely one for prospective democracies. 

Democracies, it seems, do not go to war with each other, having established a zone of peace amongst all others they perceive to be of a democratic nature as well. Acknowledging a small data set, there have only been dubious and marginal cases that would come close to disproving this hypothesis of a proposed democratic peace. Despite disagreement between rival theorists, it does not seem that democracies extend this zone of peace to the remainder of the world following nondemocratic regime types. Also, countries undergoing democratisation may in fact be more prone to violence. Limiting the scope of the word ‘fight’ provides a solid argument that democracies do not fight each other, however if one were to widen the scope to include events such as subterfuge, subversion, espionage, and militarised interstate disputes, any aura of utopian peace between democracies vanishes. Democracies as an international majority and an emerging international norm of the correct type of governance is relatively recent, and stable democracies are overwhelmingly concentrated in the West, or pro-West states. This provides ammunition for arguments that democratic rule isn’t the primary factor for the peace of these states, but is countered by empirical studies of the like of Maoz’s showing otherwise. Does that mean Immanuel Kant is right about a perpetual peace amongst republics that could eventually encompass a newly pacified world? Perhaps not especially seeing misalignments in Kant’s views of a republic and contemporary views of a liberal democracy. Regardless, if past, present, and much theory is correct, a successfully democratising world will likely be more peaceful, albeit with greater risks for violence during many a democratising process. The domestic norms of democracy prove to be well-adjusted to avoiding violence and escalation even in an international sphere, and democratic institutions help prevent rash action by leaders. Despite this, the blanket statement that all publics will universally refrain from all wars is judged incorrect, although democratic institutions and norms help prevent escalation. Whilst the evidence and theory points towards a miniscule chance of inter-democratic war occuring if it ever did we may see that Winston Churchill’s warning that “the wars of peoples will be more terrible than those of kings” rings true (International Churchill Society, 2020, online).



Becker, J., 2015. Willingly to War: Public Response to the Outbreak of War. Online. Accessed 21/12/20.

Bueno de Mesquita, B., and Lalman, D., 1986. Reason and War. American Political Science Review, Vol. 79, Issue 1.

Chan, S., 1997. In Search of Democratic Peace: Problems and Promise. International Studies Quarterly, Vol. 41, Issue 1.

Choi, A., 2003. The Power of Democratic Competition. International Security, Vol. 28, Issue 1.

Clements, J., 2012. Mannerheim: President, Soldier, Spy. Haus Publishing.

Desilver, D., 2019. Despite Global Concerns About Democracy, More Than Half of Countries Are Democratic. Online.

Accessed 20/12/2020.

Geva, N., DeRouen, K., and Mintz, A., 1993. The Political Incentive Explanation of “Democratic Peace”: Evidence from Experimental Research. International Interactions, Vol. 18, Issue 1.

Gleijeses, P., 1991. Shattered Hope: The Guatemalan Revolution and the United States, 1944-1954. Princeton University Press.

Immerman, R., H., 1982. The CIA in Guatemala: The Foreign Policy of Intervention. University of Texas Press.

International Churchill Society, 2020. Army Reform. Online. Accessed 23/12/20.

Kant, I., 1795. Perpetual Peace; A Philosophical Essay. London: S. Sonnenschein, 1903.

Kissinger, H., 1994. Diplomacy. Pocket Books: Reprinted Edition

Layne, C., 1994. The Myth of the Democratic Peace. International Security, Vol.19, Issue 2.

Mansfield, E., and Snyder, J., 2007. Electing to Fight: Why Emerging Democracies go to War. MIT Press; New Ed Edn.

Maoz, Z., and Abdolali, N., 1989. Regime Types and International Conflict, 1816-1976. Journal of Conflict Resolution, Vol. 33, Issue 1.

Morgan, C., and Campbell, S,. 1991. Domestic Structure, Decisional Constraints, and War: So Why Kant Democracies Fight? Journal of Conflict Resolution, Vol. 35, Issue 1.

Nohlen, D., 2005. Elections in the Americas: A data handbook, Volume II. OUP Oxford.

Owen, J., 1994. How Liberalism Produces Democratic Peace. International Security, Vol. 19, Issue 2.

Paul, T., 2005. The India-Pakistan Conflict: An Enduring Rivalry. Cambridge University Press.

Ray, J.L., 1998. Does Democracy Cause Peace? Annual review of political science, Vol. 1, Issue 1.

Ray, J.L, 1995. Democracy and International Conflict: An Evaluation of the Democratic Peace Proposition. Colombia: University of South Carolina Press

Reiter, D., and Stam III, A., 1998. Democracy, War Initiation, and Victory. The American Political Science Review, Vol. 92, Issue 2.

Richardson, L., 1960. Statistics of Deadly Quarrels. Boxwood Press.

Russett, B., 1993. Can a Democratic Peace Be Built? International Interactions, Vol. 18, Issue 1.

Starr, H., 1992. Democracy and War: Choice, Learning, and Security Communities. Journal of Peace, Vol. 29, Issue 1.

Systemic Peace, 2020. Polity IV Regime Trends: Lebanon 1946-2013. Online. Accessed 22/12/2020.

Weart, S., 1998. Never at War. Yale University Press.

Weart, S., 1994. Peace among Democratic and Oligarchic Republics. Journal of Peace Research, Vol. 31, Issue 1.

Widmaier, W., 2005. The Democratic Peace is What States Make of It: A Constructivist Analysis of the US-Indian ‘Near-Miss’ in the 1971 South Asian Crisis. European Journal of International Relations, Vol. 11, Issue 3.

Zakaria, F., 1997. The Rise of Illiberal Democracy. Foreign Affairs, Vol. 76, Issue 6.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s