Effective defense

This essay is on Wikiversity to encourage a wide discussion of the issues it raises moderated by the Wikimedia rules that invite contributors to “be bold but not reckless,” contributing revisions written from a neutral point of view, citing credible sources -- and raising other questions and concerns on the associated '“Discuss”' page.

Win or lose, nonviolence builds democracy,

while violence perpetuates tyranny,

on average, in the long run.

This claim summarizes the key findings of research led by Chenoweth and Stephan. Their research is outlined below along with other research on the effectiveness of alternative approaches to conflict.

These and related findings (discussed below) may ultimately lead to a radical shift in how people all over the world defend themselves and their interests. U.S. President Obama "asked the C.I.A. to analyze examples of America financing and supplying arms to an insurgency in a country that actually worked out well. And they couldn’t come up with much." The public might be better served if (a) more people asked questions like this, and (b) answers were compiled in a database subject to public review like Wikiversity.

The need for such analysis is supported by the claims of excessive power of arms manufacturers over the political and foreign policies of major nations including the United States: Governmental powers have reportedly been used in secret to pressure foreign leaders to purchase arms they didn't need, and arms sold in secret too often fall into the hands of one's enemies. Those arms sales were often accompanied by substantial bribes and subsidized by taxpayers in the arms producing countries. Many of those arms have allegedly been used to commit mass murder, e.g., for raw materials in Liberia, to name only one.

One of the most important research questions today may be to characterize what motivates people to leave the sidelines to involve themselves in conflict or to defect or modify their level of support for one side or another.

Moreover, such research should be funded and managed in a way that is clearly independent of people and organizations with a potential stake in the research results (e.g., arms producers and military or security organizations).

If the analyses below are validated in other research, it suggests that lethal force may be counterproductive except when legally constituted and widely supported security forces act to reduce a clear and present threat to the lives and safety of others.

Nonviolent resistance has been more successful than violence

Chenoweth and Stephan developed a database of all the major governmental change efforts of the twentieth century. They identified 217 movements that were predominantly violent and 106 that were primarily nonviolent. Outcomes were classified as either (1) failure, (2) partial success or (3) success. The basic results are summarized in Table 1: Nonviolence was twice as likely to achieve success as violence.

Table 1. Major governmental change efforts of the twentieth century by dominant nature of the struggle (violent or nonviolent) and by outcome (failure, partial success, success) in the NAVCO1.1 data set compiled by Chenoweth and Stephan.

	Number of conflicts		Percent(*)
	violent	nonviolent	violent	nonviolent
Outcome
success	55	57	25%	54%
partial success	28	26	13%	25%
failure	134	23	62%	22%
total	217	106	100%	100%(*)
(*) Percent within conflicts of the same primary nature. Thus, the "violent" column percents add to 100. The nonviolent total differs from 100 only because of round-off.

Nonviolence builds democracy; violence doesn't

However, the benefits of nonviolence over violence extend beyond the end of a conflict. Chenoweth and Stephan merged their data with the Polity IV database, which 'contains coded annual information on the level of democracy for all independent states with greater than 500,000 total population and covers the years 1800–2013. ... For each year and country, a "Polity Score" is determined, which ranges from -10 to +10, with -10 to -6 corresponding to autocracies, -5 to 5 corresponding to anocracies, and 6 to 10 to democracies.'

Polity Score ranges from -10 to +10

	minimum value	maximum value
autocracies	-10	-6
anocracies	-5	5
democracies	6	10

Table 2 shows the average increase in democratization from one year before the start of a conflict to one year after. The results suggest that win or lose, nonviolence tends on average to be followed by an increase in the Polity IV rating while violence has essentially no impact on democratization. As noted above, nonviolence builds democracy, while violence perpetuates tyranny, on average, in the long run.

Table 2. Average increase in Polity score from one year before to one year after a conflict. None of the changes following violent campaigns are statistically significant while all the changes following nonviolent campaigns are highly significant with significance probabilities less than 0.001.^(*)

	violent	nonviolent
success	0.5	5.9
partial success	1.4	4.2
failure	0.4	3.0
(*) Similar analyses of changes 5 and 10 years after a conflict produced similar results with the long term changes following violent campaigns actually being negative but not statistically significant.

The reality is more complicated than the simple summary of Table 2: A primary determinant of the level of democracy after a conflict, apart from the primary (violent or nonviolent) nature of the conflict, is the level of democracy before. The accompanying figures plot the Polity IV democracy score 1, 5, and 10 years after each conflict vs. 1 year before.

Democratization 1 year after (vertical scale) vs. 1 year before (horizontal scale) the end of twentieth century revolutions

Each plot is split into six panels by the primary nature (violent or nonviolent) of the conflict and the outcome (failure, partial success, success). Points on the dotted diagonal line in each panel indicate conflicts that were accompanied by zero change in their Polity scores for the indicated time frame.

Democratization 5 years after (vertical scale) vs. 1 year before (horizontal scale) the end of twentieth century revolutions

The solid lines in each panel are based on the best fit of several models considered. This expresses the democracy score after the conflict as linearly dependent on the democracy score before plus interactions between outcome and both the democracy score before and the primary nature of the conflict. For more detail, see Chenoweth and Stephan (2011).

Democratization 10 years after (vertical scale) vs. 1 year before (horizontal scale) the end of twentieth century revolutions

These plots show more detail behind the simple summary of Table 2: Successful nonviolent revolutions have on average had a substantial impact in increasing the level of democracy among autocracies but no impact among the best democracies. By contrast, the worst long term outcomes tend to be from successful violent revolutions. This is worth repeating: Successful violent revolutions provide the worst prospects on average for democracy. This has been explained by observing that successful violence brings to power people who know how to use violence but are not as good at solving problems without violence.

In sum, the overall image supports the claim made above: Win or lose, nonviolence builds democracy, while violence perpetuates tyranny, on average, in the long run.

Multiple strategies and Chenoweth's 3.5 percent rule

Chenoweth and Stephan further note that no campaign using only one strategy succeeded. They said that, "Campaigns that constantly update their information, adapt to conditions, and outmaneuver the adversary are more likely to succeed than campaigns that expect to succeed merely by virtue of their causes and methods."

Moreover, every campaign that achieved the support of 3.5 percent of the population was successful.

Effective use of air power

Robert Pape (1996) Bombing to Win: Air Power and Coercion in War (Cornell U. Pr.) compared air power supporting ground operations with strategic bombing of infrastructure (including production facilities). He concluded that air support of ground operations could be effective in defeating an enemy, but strategic bombing was a waste.

One example was the 1940-41 London Blitz by which Hitler attempted to bomb the English citizenry into supporting him. It backfired. Before Hitler started bombing London, he had destroyed most of the Royal Air Force on the ground. The British public didn't like having their air force destroyed. However, the pain of World War I was still vivid in the memories of anyone over 30, and few British citizens were eager to repeat that experience. After Hitler started bombing London, the conflict became very personal. The vast majority of British citizens supported Churchill's commitment that, "We shall never surrender."

Pape claims that the results of other strategic bombing campaigns have been similar. He and others have argued that the nuclear attacks on Hiroshima (August 6, 1945) and Nagasaki (August 9, 1945) played a minor role in Emperor Hirohito's decision to surrender (August 14, 1945). More important was the speed of the collapse of the Japanese army in Manchuria after the Soviets attacked on August 9, hours before the nuclear destruction of Nagasaki. Emperor Hirohito mentioned the atomic bombs in his surrender message to the Japanese people on August 14. However, in his surrender message to the military 3 days later, he mentioned the Soviet invasion, not the atom bombs. The reality of Hirohito's motives in surrendering cannot be determined from the historical record about that event. It helps to compare that with similar decisions by others in crudely similar situations, as Pape attempted to do.

Pape's argument against strategic bombing rests on two claims: First, strategic bombing often hardens the will to resist of civilians impacted. Second, infrastructure destroyed is usually replaced by resources harder to destroy.

Pape's analysis was questioned by a 1999 RAND Corporation study funded by the United States Air Force. A subsequent multivariate probit analysis of the RAND data supported Pape's assessment.

More recently Jeremy Scahill claims that the US use of drones in Iraq, Afghanistan, Pakistan, Yemen and elsewhere have manufactured enemies faster than they were neutralized. This seems consistent with Pape's claims in Bombing to Win.

The discussion of the Islamic State of Iraq and the Levant (ISIL) in late 2014 suggests that US military operations in Iraq and neighboring countries, especially since 2003, may have been counterproductive. This seems to support the claims of Scahill and Pape. William Astore, history professor and retired Lt. Col., United States Air Force, suggests that the record of U.S. actions since at least 2003 has demonstrated a huge capacity for harm with little evidence of an ability to achieve anything positive. He suggests that the U.S. should do what it did in Vietnam in 1975: Just leave. The locals will sort out their problems, and the U.S. can reconnect in a couple of decades.

What motivates participants in conflict?

One simple principle seems to explain the results described above:

When people are killed and property destroyed,

the apparent perpetrators often make enemies.

This might seem obvious. However, when people feel threatened, they often respond with violence in ways that are ultimately counterproductive. More research is needed to understand the evolution of conflict: Why do people leave the sidelines to support one side or the other? Why do people already supporting one side increase or decrease their level of support? Why do some defect?

A partial answer to this is provided by the research led by Danial Kahneman, which documents how people make decisions on what comes most readily to mind and often discount or don't even see evidence that contradicts their preconceptions.

Thus, collateral damage that they commit proves to us that they are subhuman or at best criminally misled. Meanwhile, collateral damage that we commit is unfortunate but necessary from our perspective but proves to them that we are subhuman or at best criminally misled.

Cheneoweth, Stephan and others provide research that helps explain people's responses to nonviolence as well as violence: The risks and level of commitment required to support nonviolence tend to be less than for violence. People are also generally repulsed by violence perceived to be excessive, especially against nonviolent protesters. Authorities sometimes recognize this by using agents provocateur to create events that can then be used to justify repression.

Conclusions

More research is needed to improve human understanding of the evolution of group identity and faultlines in conflict. Such research could help observers and participants in conflict craft more effective responses to specific challenges. It could also help anyone considering military options select ones that are more effective both in ending current hostilities and in reducing the risks of future violence.

There has been much discussion of the need to win hearts and minds in a conflict. However, there has been little discussion of how to do that. The results from Vietnam, Iraq, and elsewhere suggest that it's hard to win people's hearts and minds by killing them.

References

• Erica Chenoweth; Maria J. Stephan (2011). Why Civil Resistance Works: The Strategic Logic of Nonviolent Conflict (in en). Columbia University Press. Wikidata Q88725216. ISBN 978-0-231-15683-7.