HomeDocuments on SitePersonalDemocratic PeaceDemocide20th C. DemocideMegamurderersLesser MurderersWhy DemocideDimensionsConflictMethodsTheoryPolicyLinks PHOTOS OF DEMOCIDEGalleries

Power Kills

Contents | Preface

Chapter 1: Introduction and Summary

Appendix 1.1: Q and A on No Wars Between Democracies

Chapter 8: On The Nature of Democracy

Chapter 13: Why Does Power Kill?

Other Democratic Peace Documents On This Site


What is the "democratic peace"?

"Waging denuclearization and social justice through democracy"

"The rule of law: towards eliminating war"

"Freedom of the press--A Way to Global Peace"

"Convocation Speech"

Freeman Interview

City Times Interview

But What About...?


"The democratic peace: a new idea?"

Bibliography on democracy and war


"Libertarianism and International Violence"

"Libertarianism, Violence Within States, and the Polarity Principle"

"Libertarian Propositions on Violence Within and Between Nations: A Test Against Published Research Results"

"Democracies ARE less warlike than other regimes"


Vol. 2: The Conflict Helix (see Chapter 35)

Vol. 4: War, Power, Peace (see e.g., Propositions 16.11 and 16.27

Statistics of Democide

The Miracle That Is Freedom

Power Kills


Appendix 1.1

Q And A On The Fact
That Democracies Do Not
Make War On Each Other*

By R.J. Rummel

In early 1994, under the title "The Most Important Fact of Our Time," I posted on several internet news groups and e-mail lists the finding that democracies do not make war on each other, and suggested that through democratic freedom we now have a solution to war. This posting stimulated many questions and arguments. I then summarized the most important of these, provided my answers, and posted it on the internet under the title given this appendix. This appendix is a revision of that posting. The research details underlying this appendix are given in Chapter 2.**

Q: How is democracy defined?

A: By democracy is meant liberal democracy, where those who hold power are elected in competitive elections with a secret ballot and wide franchise (loosely understood as including at least 2/3rds of adult males); where there is freedom of speech, religion, and organization; and a constitutional framework of law to which the government is subordinate and that guarantees equal rights.***

A list of current liberal democracies includes Andorra, Argentina, Australia, Austria, Bahamas, Barbados, Belgium, Belize, Benin, Bolivia, Botswana, Bulgaria, Canada, Cape Verde, Chile, Costa Rica, Cyprus (Greek), Czech Republic, Denmark, Dominica, Ecuador, Estonia, Finland, France, Germany, Greece, Grenada, Guyana, Hungary, Iceland, Ireland, Israel, Italy, Jamaica, Japan, Kiribati, Korea (South), Latvia, Liechtenstein, Lithuania, Luxembourg, Malawi, Malta, Marshall Islands, , Mauritius, Micronesia, Monaco, Mongolia, Namibia, Nauru, Netherlands, New Zealand, Norway, Palau, Panama, Poland, Portugal, St. Kitts and Nevis, St. Lucia, St. Vincent and the Grenadines, San Marino, Sao Tome and Principe, Slovakia, Slovenia, Solomon Islands, Spain, , South Africa, Sweden, Switzerland, Trinidad and Tobago, Tuvalu, United Kingdom, United States, Uruguay, Vanuatu, and Western Samoa.

The list would be different, of course, for previous decades. For certain years of the 18th century, for example, it would include the Swiss Cantons, French Republic, and United States; for certain years during 1800-1850 it would include the Swiss Confederation, United States, France, Belgium, Great Britain, Netherlands, Piedmont, and Denmark.

Q: But can you really apply the contemporary definition of democracy to previous centuries?

A: There is a problem and it resides in how far equal rights and the franchise is extended, as before women achieved equal rights, the franchise was limited to property owning males, or slavery existed. For previous centuries the definition of democracy has been loosened be researchers to include at least, as mentioned, two-thirds males having equal rights (as long as the lower classes were not excluded), while maintaining the other characteristics (equal rights, open competitive elections, etc.). For one, democracies so defined in previous centuries, such as the United States in 1800 and democratic classical Athens, saw themselves as democratic, called themselves democratic, and were perceived by other nations as democratic. Second, even with this looser definition, well established democracies so defined still did not make war on each other. Well established means that a regime had been democratic long enough for it to be stable and democratic practices to become established.

The fundamental question about any definition is: does it work? Does it define something in reality that predicts systematically to something else. If we have so defined an x such that it regularly predicts to y, then that is a useful and important definition of x. In the definition I have given above of democracy it predicts to a condition of continuous peace (nonwar) between nations defined as democratic. If one does not agree that these are democracies, fine. Then call them xcracies. We then still can say that xcracies do not make war on each other and by universalizing xcracies we have a solution to war.

Q: Still, is not the historical sample of democracies too small for such a broad generalization.?

Whether the definition of democracy is broad or narrow, we have statistical means to calculating if the number of democracies is in fact significant (the same kind of statistics medical researchers use to test the significance of drugs or symptoms). That there have been no wars between democracies since, say, 1816, is statistically significant. That is, given the historical number of democracies, the probability of the hypothesis that democracies have never made war on each other being wrong is very low (given the consistency of findings across diverse studies, the odds must be surely millions to one).

Q: But are there not other factors really accounting for the lack of war between democracies, such as geographic distance?

A number of studies of whether democracies made war on each other have tried to determine if there is a hidden factor accounting for this, such as economic development, industrialization, geographic distance, trade, alliances, and so on. Always, democracy comes out as the best explanation.

Best is meant in a statistically significant sense. That is, the probability that democracy would not be a determinant when these other factors are considered is very low (odds also of tens of thousands to one). This has been gauged through analysis of variance and various kinds of regression analysis.

Q: Cannot statistics be used to prove anything?

True, statistics can be misused and have been, but this is true of any scientific method. Virtually all the medical drugs one takes today are based on statistical tests, not unlike those used to test whether democracies do not make war on each other is a chance occurrence. If one is going to be cynical about statistics, then one should also be very wary of taking any modern drugs for an illness or disease. This issue is really not statistics but how well they have been applied and whether the data meet the assumptions of the statistical model used.

Now, some statistics. If one defines an international war as any military engagements in which 1,000 or more were killed, then 353 pairs of nations (e.g., Germany vs. USSR) engaged in such wars, 1816-1991. None were between two democracies, 155 pairs involved a democracy and a nondemocracy, and 198 involved two nondemocracies fighting each other. The average length of war between states was 35 months, average battle deaths was 15,069.

A good way of calculating the statistical significance of democracies not making war on each other is through the binomial theorem. For both one requires several statistics: the number of nondemocratic pairs and democratic pairs of states in the world for the period during which the wars between these types of pairs occurred, and the number of wars between each type. The problem has not been in determining the number of democratic pairs, but how many nondemocratic pairs there are for some period of time. This has been confronted in the literature, and for those periods in which this number could be defined the zero wars between democracies has been very significant (usually much less than a probability of .01 that this zero was by chance). Just one example follows.

For the years 1946-1986, when there were the most democracies and thus the hardest test of the proposition that democracies do not make war on each other, there were over this period 45 states that had a democratic regime; 109 that did not. There were thus 6,876 state dyads (e.g., Bolivia-Chile), of which 990 were democratic-democratic dyads, none of which fought each other. Thirty-two nondemocratic dyads engaged in war. Thus the probability of any dyad engaging in war 1946-1986 was 32/6876 = .0047; of not engaging in war is .9953. Now, what is the probability of the 990 dyads not engaging in war during this period? Using the binomial theorem, it is .9953 to the 990th power = .0099, or rounded off .01. This is highly significant. The odds of this lack of war between democracies being by chance is virtually 100 to 1.

One should not take this result in isolation, since the lack of war has been tested in different ways for other periods, definitions of democracy, and ways of defining war, and in each case has been significant. Thus, the overall significance is really a multiple (or function, if some of these studies are not independent) of these different significant probabilities, which would make the overall probability (subjectively estimated) of the results being by chance alone surely at least a million to one.

Q: But your statistics are for the Cold War period. Was not the lack of war between democracies really due to the threat of the Soviet Union?

My above test for the years 1946-86 is not the only one. As mentioned, other tests have been done for different years, including 1816-1960. Now it may be true that the Cold War accounted for the particular lack of war between democracies, but what about other periods. Also, ignore the statistics and consider Europe, the historical cauldron of war, and what has happened since the end of the Cold War. Unity has continued to grow, rather then hostility. And, incredible, those old enemies, France and Germany, have even considered forming a common army. Moreover, once the former enemies became democratic, they have tried to join and are being integrated into a larger Europe.

Q: But can you meet the assumptions of the statistical model, particularly that of randomness?

All statistical tests on humans suffer from the inability to truly meet the assumption of randomness (equal likelihood of each case, event, sample point) basic to the model. In medical tests, whether double-blind or not, the sample is usually constrained to Americans, students, doctors, etc., and thus may introduce unknown masking factors. Even ignoring this, any statistical test is only giving results in terms of probabilities, and for that one test the improbable may in fact have occurred. This is why no researcher should accept any one or two tests as definitive. Only if a range of tests are consistent over many kinds of data, researchers, and methods can one have confidence in the results. This is true for vitamin E supplements reduces the risk of cancer; it is true for the proposition that democracies do not make war on each other.

Q: But such is raw empiricism. You must have theory. Where is your theory for democracies not making war on each other?

The first theory for this goes back to Immanuel Kant's Perpetual Peace published in 1795, way before any related empirical research. Kant's theory (which is presented in Chapter 10) has been elaborated and absorbed into the modern explanation, which is that:

Q: But what about the American Civil War, War of 1812, Spanish-American War, democratically appointed Hitler and WWII, Finland versus the Allies in WWII, current war in former Yugoslavia, etc. etc.? Did not these constitute wars among and between democracies?

Those who have investigated these and many other exceptions (and especially Ray [1995] and Weart [1995]) have concluded that these do not constitute meaningful exceptions. The United Kingdom versus Finland is perhaps the most often mentioned, next to the war of 1812, as a possible exception. Although the United Kingdom did bomb German run mining operations in Finland, there was no actual fighting between Finnish forces and those of the democratic allies. Regarding the war of 1812, the United Kingdom was not a democracy by any definition until later in the century. Regarding Hitler, once he was given the power to rule by decree in 1933 and suppressed opposition, his government was no longer democratic. Freedom of speech and religion, along with other rights, was eliminated; regular competitive elections were no longer held, and the Nazis were above the law.

Regarding the American Civil War, an often mentioned exception, the South was not a sovereign democracy at that time. For one, it was not recognized by any major Power, which means that it was not recognized as an independent state. But aside from this, the franchise was limited to free males (which constituted about 35 to 40 percent of all males in the Confederacy), President Jefferson Davis was not elected, but appointed by representatives themselves selected by the confederate states. There was an election in 1861, but it was not competitive.

As with many facts by which we guide our lives, we need not be hung up on such possible exceptions. All alleged exceptions are at the margins of what we call liberal democracies. Although none have been accepted as exceptions to the rule by those who have done research on them, let us suppose that they are in fact exceptions. This still would not weaken the proposition that well established democracies do not make war on each other. This is because in no case have undoubted democracies (such as Sweden, Norway, Belgium, France, United States, and Canada) made war on each other and none are mentioned as exceptions.

Q: What are undoubted or well-established democracies?

A well-established democracy is one for which enough time has passed since its inception for peace-sufficient democratic procedures to become accepted and democratic culture to settle in. Around three years seems to be enough for this.

Then what is an undoubted democracy? It is what all who write on liberal democracy would recognize as clearly a democratic country; it is what all published definitions of democracy would include as a democracy; it is what all scales of democracy based on operational definitions measure as a democracy. These would include such nations as Denmark, Sweden, Canada, the United States, France, Great Britain, and the like. They all have in common several characteristics: their citizens regardless of class have equal rights; policies and leaders are determined through open and competitive elections and voting; and there is freedom of speech. One can be precise about these characteristics and related ones, scale them, weigh and sum them some way, and thus measure democracy and array nations on a scale of democracy. This has been done by a number of researchers, including those who have tested whether democracies make war on each other.

Q: Okay, okay, what about democracies, particularly the United States, carrying on covert action against other countries, some of which were democracies?

Yes, but this was during the Cold War and was part of the largely successful policy to contain communism, particularly Soviet power. Mistakes were made, actions were taken which in hindsight many democrats are embarrassed about. Even then, there was no military action between democracies.

This having been said, there is also a deeper explanation. Democracies are not monolithic; they are divided into many agencies, some of which operate in secrecy and are really totalitarian subsystems connected only at the top to democratic processes. The military, especially in wartime, and the secret services, such as the CIA, are examples. These near isolated islands of power operate as democratic theory would assume. Outside of the democratic sunshine and processes, they do things that were they subject to democratic scrutiny would be forbidden. The answer to this problem is more democratic control. And with the spread of democracy around the world, armies and secret services would be less and less needed. Indeed, with near universal democratization, they could be eliminated altogether.

Q: Even if that democracies do not make war on each other is true, how can you generalize to the future? Because something never happened in the past you cannot say it will not in the future.

That democracies do not make war on each other, that they create a zone of peace among themselves, is now the most firmly established proposition in international relations and the most important. Given this, we have a solid base for forecasting that there never will be a war between democracies and that universalizing democracy will end international wars.

All public policies are based on perceptions of historical patterns. Indeed, all scientific predictions are based on established theoretical/empirical patterns. No prediction of the future is thus certain; all are based on the past. The question is how good the established patterns are that underlie the predictions. Are they reliable, well verified, theoretically understood. The historical pattern that there is no war between democracies meets all these requirements. Even those who have been very skeptical when starting their research on this have become convinced. One has said that this is now the best established law of international relations.

Given all this and the absolute importance of eliminating war, should we not implement the best empirical/theoretical solution now in our hands? That is, as practical and desired by the people involved, to universalize democracy?

Q: Yes, yes, but how can YOU be so sure?

No one can be certain. There is always the possibility that one can be wrong in fact and theory. But I am confident enough that democracies do not make war on each other to believe that it is the best solution to war. I started research on war and peace in 1956 and have spent a professional research career in political science on it since. The whole character of this life time of research supports the proposition. But others have done their own research and come to the same conclusion. And like others I have gone from unbelieving (its too simple and simplistic) to a maybe-but, to full acceptance as the number of positive research studies and theoretical elaborations have accumulated.

Q: Since everyone is in favor of democracy anyway why make a big thing of this?

Because it will take the investment of much resources by the United States and other democracies to help nations democratize. Russian alone needs tens of billions in aid to further democratization. Such aid will be more forthcoming and more broadly supported if there is a wider understanding among the democracies that by providing human and financial resources to democratize we are not only promoting the freedom and prosperity of other countries but also peace and nonviolence. Such aid is cheap compared to the likely human and material cost of future wars.

There is also the struggle for human rights in many countries. It helps the struggle to be to not only justify human rights for their own sake, but to point out their importance for global peace and security.

Q: You keep mentioning research. What research on this has been published?

See the citations in Chapter 2**. 


* From the pre-publisher edited manuscript of the "Appendix to Chapter 1" in R.J. Rummel, Power Kills: Democracy as a Method of Nonviolence, 1997. For full reference to Power Kills, the list of its contents, figures, and tables, and the text of its preface, click book.

** See the table of contents. [since Chapter 2 is not on this web site, see the annotated bibliography and bibliography on peace and war to The Miracle That Is Freedom]

*** On the nature of democracy, see Chapter 8.

For citations see the Statistics of Democide REFERENCES

You are the visitor since 11/23/02

Go to top of document