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

Review of Power Kills

Rummel's reply

Ted Galen Carpenter's rebuttal to Rummel's reply

H-C. Peterson's response to Carpenter' review and rebuttal.

Other Related 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


Bibliography on Democracy and War

Q & A On Democracies Not Making War on Each Other

But What About...?

"The democratic peace: a new idea?"


"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


Democracy And War *

By Ted Galen Carpenter

R. J. Rummel, professor of political science at the University of Hawaii, has justifiably acquired a reputation as an outstanding scholar of violence perpetrated by the political state. His book Death By Government (New Brunswick, N.J.: Transaction Publishers, 1994) is a detailed, searing, and compelling indictment of the mass murder (more than 169 million victims) committed by governments during the twentieth century.

Rummel has also long been a proponent of the "peaceful democracies" thesis: that democracies are markedly less prone than are authoritarian or totalitarian states to resort to violence in the conduct of their external relations and that democracies never (or almost never) fight other democracies. In Power Kills: Democracy as a Method of Nonviolence (New Brunswick, N.J.: Transaction Publishers, 1997), he provides the most systematic development of that thesis to date. Unfortunately, his reach greatly exceeds his grasp.

The realist faction of foreign policy scholars--especially the so-called structural realists, who argue that conflict is inherent in an international system that has no central authority and is made up of nation-states with conflicting interests--is Rummel's principal target. The realists are wrong across the board, he contends.

Wrong with regard to war and lesser international violence. Wrong about civil collective violence. Wrong about genocide and mass murder. There is one solution to each and the solution to each case is the same. It is to foster democratic freedom and to democratize coercive power and force. That is, mass killing and mass murder carried out by government is a result of indiscriminate, irresponsible Power at the Center.
----(p. 3)

In addition to that general conclusion, Rummel asserts that several key propositions about democracy and violence have been "uncovered or verified." Among the more important are the following: "Well established democracies do not make war on and rarely commit lesser violence against each other" (p. 4). "The more two nations are democratic, the less likely war or lesser violence between them" (p. 5). "The more a nation is democratic, the less severe its overall foreign violence" (p. 5). Indeed, Rummel contends that for "theoretical reasons" he "would expect no violence between democracies at all" (p. 101).

He advances (with varying degrees of conviction) three explanations for the peaceful nature of democracies. The first-levcl explanation is that the publics in democratic societies generally prefer to avoid war. Rummel argues that although that factor (emphasized by other scholars) has some validity, it must be viewed with caution. He acknowledges that "democratic peoples have become jingoistic on occasions and enthusiastically favored war.... They can also be aggressive today, pacific tomorrow" (p. 132).

Rummel attaches greater importance to the second-level explanation: the influence of democratic institutions and culture. "Where by virtue of their institutions democratic people must, to maintain democracy, negotiate and compromise rather than fight, this becomes part of the cultural heritage" (p. 138).

Moreover, he states,

since we deal with others through a cultural matrix, it is also natural for democratic people to perceive other regimes in these terms, to believe that all basic issues between nations can be settled by people sitting down at a table and talking them out, and to tolerate the existence of other regimes and ideologies that do not openly threaten one's democratic way of life.
----(p. 138)

The converse is equally true: totalitarian regimes see other regimes as being as ruthless, duplicitous, and brutal as themselves, and they act accordingly, thereby intensifying the cycle of violence.

Even more significant than the impact of democratic political culture, Rummel contends, is the third-level explanation: the operation of a "social field" based on diversity and individual freedom. "This spontaneous social field of constantly interacting individuals and groups, all pursuing their own interests, is a field of continuous nonviolent conflict" (p. 165; emphasis in original). In other words, the way to minimize violence, both domestic and international, is to decentralize power by strengthening civil society and constraining the role of the state.

It is a truism of science (even social science) that extraordinary claims require extraordinary evidence. Because Rummel affirms not only that democracies are on balance more peaceful than authoritarian and totalitarian systems but that a global system consisting entirely of democracies would produce a world without war, he faces a daunting burden of proof. In Power Kills, he fails to provide the necessary evidence.

Rummel does a credible job of making the case that there is a continuum of violence: as one moves from democratic states to authoritarian and then totalitarian ones, the level of violence, both domestic and international, increases. Even on that point, however, both his methodology and his arguments are sometimes dubious. For example, he attaches great weight to the fact that historically the less democratic a regime, the higher the number of battle deaths. Such data are designed to show both that democratic governments are less inclined to put their people through the meat grinder of war and that democratic populations and institutions are less tolerant of battlefield casualties.

Both propositions may well be true, but battle-death figures hardly provide compelling evidence. There are several possible alternative explanations for the markedly lower battlefield fatalities among democratic states. Most obvious, democracies have been on the winning side in most wars during the twentieth century--an important consideration because the losing side typically suffers disproportionately. The meager number of U.S. fatalities in the Persian Gulf War, for example, would appear to have had more to do with the superiority of American military technology and the abysmally stupid strategy of Saddam Hussein (a static defense in open desert terrain) than with the virtues of American democracy. Similarly, military superiority was the most probable reason that U.S. forces inflicted far more casualties than they incurred in the Korean and Vietnam conflicts.

Equally questionable are Rummel's assertions that wars between democracies (if such events occurred at all) would be less violent than those between authoritarian states and that wars between totalitarian states are, and will inevitably be, the most violent. Again, other factors can influence, and perhaps even determine, the magnitude of warfare. If the Cold War between the United States and the Soviet Union had ever turned hot, the struggle involving such a mixed dyad (one democratic state and one totalitarian state) would have been enormously destructive. Indeed, it would have been far worse than the war between China and Vietnam in 1979, even though both belligerents were totalitarian. The reason is self-evident: the United States and the Soviet Union had huge, capable military forces (including thousands of nuclear weapons). By comparison, Chinese and Vietnamese military forces were relatively puny in their destructive capacities.

Such examples illustrate a more general failing in Power Kills. Rummel repeatedly seems oblivious to or casually dismissive of alternative explanations of the phenomena he examines. Nowhere is that tendency more evident than in his core thesis--that democracies do not wage war against other democracies.

In making that case, Rummel relies heavily on the work of other "democratic-peace" scholars, such as Bruce Russett and Michael Doyle, as well as on his own research. Unfortunately, their scholarship usually mirrors the weaknesses in Rummel's. Quoting liberally from such studies to "refute" the arguments of realists, therefore, produces a less-than-compelling case.

For example, Rummel favorably cites Doyle's observation about Italy's abrupt decision to switch alliances as World War I began.

Italy, the liberal member of the Triple Alliance with Germany and Austria, chose not to fulfill its obligations under that treaty to support its allies. Instead, Italy joined in an alliance with Britain and France, which prevented it from having to fight other liberal states and then declared war on Germany and Austria. BR>----(p. 177)

Rummel proceeds to emphasize the point that "Italy changed from the side it was obligated to fight on to line up with the democracies" (p. 177)

At the least, both scholars have engaged in egregious oversimplification. Many reasons prompted Italy's decision, but few historians would maintain that a sense of democratic solidarity was the dominant motive. Italian leaders had far more mundane considerations--most notably the desire to detach large portions of Austro-Hungarian territory in the southern Tirol and the northwestern Balkans. It is revealing that London and Paris, Rome's new democratic allies, did not merely appeal to the Italian leaders' sense of democratic solidarity; they explicitly assured the Italians that their territorial claims would be recognized in the event of an Allied victory.

That same tendency to minimize or ignore factors other than the existence of democracy as an explanation for the apparent lack of wars among democratic states appears throughout the book. Rummel is almost contemptuous of the important article in which the RAND Corporation's Christopher Layne examined a number of military "near collisions" between democratic states ("Kant or Can't: The Myth of the Democratic Peace," International Security 19 [Summer 19941: 5-49). Those included the U.S.-British confrontation over the Trent affair in 1861, the Anglo-American crisis over the Venezuela boundary dispute in the mid-1890s, and the British-French war scare over control of the Nile River (the so-called Fashoda incident) in 1898. Layne's overriding point is that in every case, the bulk of the evidence indicates that the parties pulled back from the brink of war not because of domestic pressures against fighting another democracy but because of realist strategic calculations. For example, French leaders concluded that they could not win a war against Britain, and British leaders in the 1890s concluded that Wilhelmine Germany posed a more serious long-term threat to British interests than did American preeminence in the Western Hemisphere.

Rummel dismisses Layne's treatment as subjective conjecture, emphasizing that the only pertinent fact is that in all those cases the countries involved in the crises "did not fight" (p. 42, emphasis in original). But the probable reasons for a phenomenon are at least as important as its existence. And no robotic invocation of statistical data on the alleged lack of armed combat between democracies ought to spare a scholar from the need to engage in such analysis. (After all, one can establish a strong statistical correlation between the crowing of roosters and sunrise. It would, however, be manifestly absurd to assume that the former caused the latter.)

The need to consider probable reasons for a phenomenon becomes even more apparent with another of Layne's examples, the French military occupation of Germany's Ruhr region in the early 1920s. A case in which one democratic country forcibly seized the most economically valuable portion of its democratic neighbor's territory should have elicited an extended discussion from Rummel. The mere fact that Weimar Germany did not militarily resist the occupation is hardly the salient point. The contemporary and historical records arc quite clear that German leaders recognized their country's military impotence and realized they could not prevail. To contend that the lack of a war in that situation supports the proposition that democracies do not fight other democracies borders on perversity.

Unfortunately, Rummel's failure to adequately discuss the Ruhr incident is not an aberration. Throughout the book he avoids the "hard cases" that might cast doubt on the peaceful democracies thesis. One example is the Boer War the late 1890s, a British bid to conquer the Orange Free State and the Transvaal. That conflict produced shocking acts of brutality on both sides. Because Britain was indisputably democratic (Rummel concedes that point) and the Boer states were, in his lexicon, "oligarchic republics" (politically democratic within a restricted electorate--in this case, whites only) and were perceived as democratic by the British, the war ought to be a troubling episode for Rummel and like-minded scholars. Yet he barely mentions it.

Rummel's treatment of America's War between the States is little better. In cursory discussion he states that no major power recognized the Confederacy as an independent state and he asserts that it was not a real democracy in any case because only white males could vote and President Jefferson Davis was not directly elected. The inconvenient matter that Southerners considered their new confederacy democratic (which it was by the standards of the day) and that most Northerners did not dispute that view (they merely regarded it as beside the point) is simply ignored. The willingness of democratic Americans to wage an enthusiastic internecine slaughter fairly cries out for a more serious discussion. If a democratic people could do that to their own, how confident can we be that two democracies divided by culture or race (e.g., the United States and Japan) would recoil from doing so? At the very least, proponents of the democratic-peace thesis cannot assume that the point is self-evident.

A third hard case virtually ignored by Rummel is the western front in World War I--the bloody struggle between Britain and France on one side and Wilhelmine Germany on the other. Layne and other scholars have made solid arguments that Germany was a democratic state, as it had an elected parliament with significant powers, vigorously contested elections involving multiple parties, broad suffrage, and a reasonably free press. Although it also had some autocratic features, so did Britain and France.

World War I tends to give democratic--peace theorists intellectual indigestion, and for good reason. If Wilhelmine Germany is acknowledged to have been democratic, World War I alone is probably enough to falsify the democratic-peace thesis, given the extent of the bloodletting on the western front. (To argue otherwise would risk creating the social science equivalent of the old joke "Other than that, Mrs. Lincoln, how was your evening?") Of course, Rummel may have had irrefutable evidence that Wilhelmine Germany was not democratic, but if so his readers were entitled to see it.

To have engaged that issue, however, would have required him to be far more explicit about the features of a democracy and why certain countries during the nineteenth and twentieth centuries were or were not democratic. Instead, we get a vague and slippery treatment of that topic. Frequently, it is not clear whether Rummel is applying his own definition or that of one of his democratic-peace colleagues. Nor is it always clear whether certain political features are indispensable or to what extent the definition of a democracy depends on the norms of the era being discussed. That unsystematic approach produces erratic and arbitrary designations. For example, Rummel asserts that Britain did not become a liberal democracy until 1884 (p. 109). Why was that year so important? Because the franchise was then extended to agricultural workers. But why did that measure make Britain a democracy even though women (half the adult population) were not granted the vote until several decades later?

Rummel is so determined to prove his thesis about peaceful democratic solidarity that he virtually ignores conflicting evidence. Sometimes that tendency produces embarrassing overstatements clearly at odds with the facts. He asserts at one point that "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" (p. 17; emphasis in original). But both present and former U.S. intelligence officials acknowledge that democracies routinely spied on one another, even during the Cold War, when they confronted a dangerous mutual security threat. (The Jonathan Pollard case, in which Israel conducted espionage against the United States, is merely one celebrated example.) Indeed, democratic intelligence services sometimes did more than spy. Another episode barely mentioned by Rummel was the successful CIA effort to overthrow the elected governments of Iran (1953) and Guatemala (1954). Those incidents cast further doubt on the democratic-peace thesis; if officials in the United States were willing to mount covert operations to overthrow two sister democracies--and bring highly authoritarian regimes to power in both instances--why should we assume that they would have recoiled from using military force?

Moreover, intramural democratic espionage has not abated with the end of the Cold War. As the Wall Street Journal's John Fialka, Cato Institute research fellow Stanley Kober, and others have shown, espionage (particularly economic espionage) has actually increased since the end of the Cold War--precisely the opposite of what Rummel would have predicted. Within weeks of the publication of Power Kills, U.S. Navy intelligence analyst Robert Kim was convicted of passing classified information to South Korea, not only another democracy but a U.S. military ally.

In the broadest sense, the argument that the world would be better off if all countries had strong civil societies and were governed by decentralized, democratic regimes is correct. It would almost certainly be a more peaceful world. But that is not the same as claiming that universal democracy is a panacea that will banish war. That democracies have never waged war against other democracies remains highly debatable despite the categorical assertions presented in this book. Likewise, Rummel's analysis does not establish that the absence of armed conflict between democracies since World War II (a rather brief period in any event) is due to the factors he identifies rather than other influences--for example, the existence of a powerful totalitarian threat that inhibited intrademocratic squabbles. Rummel makes an array of extraordinary claims, but he ultimately fails to prove his case.


* Scanned from Ted Galen Carpenter's "Review Essay: Democracy and War," The Independent Review 2 (Winter 1998): 435-441. Carpenter is Vice President, Defense and Foreign Policy Studies, at the Cato Institute.

You are the visitor since 11/27/02

Go to top of document