Drone strikes have been a core strategy of the so-called global war on terror. But there have also been many questions raised surrounding the effectiveness, transparency, legitimacy, and ethics of their use.

Technology has fundamentally altered not only how security is defined, but also how it is sustained and even enhanced.  Nowhere is this new reality more apparent than in the so-called “global war on terror,” where there is little agreement about counterterrorism tactics and strategy.  A core part of the so-called war on terror has been the utilization of drone technology.  In the context of warfare, the drone has at least three functions: surveillance, killing, and providing targeting for another weapons system. The significance of the new technology is not so much that drone operators must decide between surveillance and firing but that they can decide.  The drone often removes the need for indirect fire (where the shooter cannot see the target).  Under such circumstances, the use of drones is a significant advantage to the side employing them.

Analysts point to several factors indicating why targeted killings by the United States (U.S.) are likely to increase in the foreseeable future. Drone strikes put fewer American lives at risk and provides a low-cost alternative to expensive and unwieldy conventional forces, especially given projected cuts in the defense budget and a dwindling public appetite for long wars.  The reasons for the shift to combat drones are obvious:  it lessens the burdens and responsibility on a state’s taxpayers, policymakers, and military.  But drones have drawbacks, too.

From a broad perspective, the use of armed drones in response to terrorism may actually be counterproductive.  It has at times proved detrimental and terrifying, not just to the targeted individuals but to entire populations, killing innocent civilians and fueling resentment that has fed into terrorist recruitment and radicalization, intensifying the very terrorism that the drones are intended to combat. Those fears have been made ever more real by the surging number of casualties caused by targeting high-value terrorists in Pakistan, Yemen, Somalia, Iraq, Syria, and Afghanistan.  The debate over the proper use of drone strikes abroad remains far from settled and has raised many questions about their effectiveness, transparency, legitimacy, and the ethics surrounding their use.  These issues deserve more attention.

Legal and Moral Issues

Image by Defence Images/Flickr

Despite frequent condemnation of the U.S. cross-border drone strikes as patently illegal, the legality question is not so straightforward because international law is not precise.  Even though the U.N. Charter explicitly prohibits states from employing “the threat or use of force against the territorial integrity or political independence of any state” (Article 2(4)), it provides two exceptions, recognizing an “inherent right of individual or collective self-defense if an armed attack occurs against a Member of the United Nations.” (Article 51).  The other exception relates to authorization by the Security Council (Articles 39, 41, 42).  Debate over the breadth of the self-defense exception dates back to the 1950s, focusing on the “inherent” nature of the right, what constitutes an “armed attack,” and when an armed attack “occurs.”  This is the essence of the current controversy over pre-emptive self-defense, which the United States invokes to justify preventing an attack by responding to it before it actually occurs.

The International Court of Justice (ICJ) has generally treated self-defense as a narrow exception to the prohibition on force.  In the 1985 case of Nicaragua v. United States, for example, it held that to give rise to a right of self-defense, an attack must be a significant one.  The ICJ has also upheld the principles of necessity and proportionality, finding that self-defense is not permissible even against an armed attack if the self-defense is not necessary to accomplish the purpose of defense or if it is disproportionate in terms of civilian lives or property lost.  Perhaps the court’s most important finding is that the prohibition on the use of force and the limited self-defense exception have become a part of customary international law.

As Rosa Brooks has argued,  ambiguity and vagueness in these core legal concepts of “self-defense” and “armed attack,” as well as related doctrines of “imminence,” “proportionality,” and “necessity,” permit the U.S. to make plausible arguments for legality, while allowing other states simultaneously to condemn the attacks as unlawful.  In the absence of a single overarching international authority and judicial system to declare who is right, the answer, if it ever arrives, will depend on the development of a consensus within the international community, which could take many years to build.

It is not even clear that use of drones against suspected terrorists is governed by the law of armed conflict (LOAC) in the first place.  If these are more appropriately regarded as law enforcement actions, as some believe, then they should be governed by law-enforcement rules and limited by international human rights law.  The intentional targeting of suspected terrorists poses vexing questions surrounding the legal principle of ‘innocent until proven guilty.’  As a matter of U.S. constitutional law and criminal law, does the executive branch, acting through the military or the intelligence community, have the right to kill a suspected terrorist whose guilt has not been adjudicated in court?  Does it violate the right to life and the prohibition of arbitrary killing, protected by, among other things, Article 6(1) of the International Covenant on Civil and Political Rights?

In U.S. law, a drone attack, like any other targeted killing, arguably, but not necessarily, violates a ban on assassination by U.S. personnel dating back to an executive order issued by President Gerald Ford in 1976. Until 1975, many high officials inside the U.S. government, including President Ford, did not know that the Central Intelligence Agency (CIA) had ever plotted to kill foreign leaders. All that changed, however, as a result of a series of exposes published in The New York Times by investigative journalist Seymour Hersh. In early 1976, following several disclosures, investigations and public revulsion, President Ford issued the executive order banning the assassinations.  The ban on assassination is still in effect in a later executive order promulgated by President Reagan.

Another question is how those who employ armed drones can justify ‘collateral damage’ to innocent bystanders who become unintended victims.  The LOAC allows the targeting of enemy combatants and expressly prohibits targeting civilians, but so long as reasonable steps are taken to avoid collateral injuries, and the loss of civilian lives is proportional to the military advantage, the accidental killing of civilians is not a war crime.  But this does not mean that it is morally or politically justified.  More fundamentally, international law raises questions about the right of the U.S. to target individuals without the consent of the government on whose territory the killing occurs.  Does the UN Charter’s Article 2(4) prohibition on the use or threat of force against the territorial integrity or political independence of another state even allow such attacks?

The U.S. has argued that the attacks are permissible because the targeted state is either unwilling or unable to police its own territory and prevent the targeted individuals from carrying out terrorist acts.  The 2005 ICJ case of Congo v. Uganda appears to weaken the U.S. argument, holding that Uganda’s military incursion into Congo to stop cross-border attacks by Congo-based insurgents was unlawful.  Most scholars and most states appear to adopt the ICJ’s broad understanding of the Article 2(4) prohibition on force and narrow understanding of Article 51’s self-defense exception.  Nevertheless, the debate continues.

Both the Bush and the Obama administrations have argued that the United States should maintain its ability to use all of the tools in its arsenal, including armed drones, to prevent terrorist organizations and groups from attacking the U.S. homeland.  On September 17, 2001, President Bush signed an executive finding that authorized the CIA to “kill or capture al-Qaeda militants around the globe.”  While some officials within the Bush administration defended the drone strikes as consistent with and conforming to international law, others emphasized their effectiveness rather than their legality, arguing that the use of drones has given the U.S. a new dimension of capability that most other nations lack.  Still, others have added that some limits must be placed on drone strikes against U.S. citizens overseas—that is, Americans should not be targeted without prior approval by a military panel or a federal judge.

On balance, the U.S. government continues to regard the drone program in Yemen, Somalia, Afghanistan, and the border regions of Pakistan as part of the ongoing U.S. war with al-Qaeda, which has been waged pursuant to the 2001 Authorization for the Use of Military Force by which Congress authorized the president to take military action against nations, organizations, or persons involved in the 9/11 attacks.  As long as the attacks are aimed at individuals associated with al-Qaeda and are for the purpose of preventing future acts of terrorism against the United States, they appear to fall within the scope of the authorization.  The U.S. government contends that international law permits the United States to use force against al-Qaeda and its affiliates in countries where there is an extant armed conflict to which al-Qaeda or its associates are party.  If the drone strikes are part of the war with al-Qaeda, the argument goes, the law of armed conflict applies. The Obama administration has taken the view that the law of armed conflict applies to drone strikes, whether they are part of the war or are used as a separate military strategy such as counterterrorism.

The ethical and legal issues raised by the rapidly developing drone technology pale in comparison to those presented by the Pentagon’s development of new autonomous weapons systems.  These amount to fully independent robots, guided by artificial intelligence, which can decide on their own whom and when to kill.  These projects, to which the Defense Department has committed billions of research dollars, have prompted an intensifying debate among legal scholars and ethicists:   “Can a machine be trusted with lethal force?”  “Who is at fault if a robot attacks a hospital or a school?”  “Is being killed by a machine a greater violation of human dignity than if the fatal blow is delivered by a human?”  A Pentagon directive requires that autonomous weapons use “appropriate levels of human judgment.”  Scientists and human rights experts have argued that the standard is far too broad, insisting that such weapons be subject to diligent application and “meaningful human control.”

 Transparency and effectiveness

Reaper Drone image by Wikimedia Commons.

Critics have argued that the U.S. drone program lacks transparency and is largely unknown to the general public and most government officials, including most members of Congress.  There is also little doubt that innocent civilians are dying in drone attacks. Some studies have demonstrated the disconnect between public statements and what researchers have discovered about civilian casualties from U.S. drone strikes. White House counterterrorism advisor John Brennan has often attributed “surgical precision” and “laser-like focus” to the drone program.  Critics argue that stressing the notion of surgical precision in the face of many civilian casualties caused by such attacks is downright misleading.  In some cases, the CIA may not even have known the identity of the people it has killed.  The presumption that all military-age males killed in drone strikes have been “militants” cannot withstand strict scrutiny.

Several organizations or publications have informed the public debate on civilian deaths from drone strikes.  These include, among others, the New America Foundation (NAF), the Long War Journal (LWJ), the Bureau of Investigative Journalism (BIJ), the Columbia Law School Human Rights Clinic (CHRC), the International Human Rights and Conflict Resolution Clinic at Stanford Law School, and the Global Justice Clinic at the NYU School of Law, which have conducted an investigation into several aspects of the U.S. targeted killing program in Pakistan and have provided a detailed narrative about the law and the policy behind it.

Despite Brennan’s and the CIA’s denials of unintended civilian deaths, the Bureau of Investigative Journalism in London has reported that 371 drone strikes in Pakistan killed between 2,564 and 3,567 people between 2004 and the first half of 2013.  Between 411 and 890 (12%-35% of the total) were civilians.  Fewer than one-quarter of those killed in drone strikes in Pakistan have been civilians. As of August 2016, President Obama has authorized and confirmed 506 drone strikes, killing an estimated 3,040 military combatants and 391 civilians.

The lack of government transparency on drone strikes raises serious questions about their effectiveness and accuracy.  If the drone attacks are to be effectively utilized, critics argue, they have to be used for short-term interventions with the intention of using them rarely, selectively, transparently, and only against those who can realistically target the United States.  Absent a realistic threat against the U.S., it is difficult to justify a killing as self-defense and thus permissible under Article 51.  Otherwise it is arguably just an extrajudicial killing of an un-convicted, often unindicted, criminal suspect as well as a violation of the U.N. Charter’s prohibition on the use of force.

 Managing Risks or Seeking Long-Term Solutions

It is time to think outside the box in which the states fighting terrorism have locked themselves, and to make the case for why the U.N. and other development organizations should be empowered and encouraged to support civic engagement, societal improvement, and low-level civil society rebuilding as a means to battle the unrest and despair that fuels terrorism.  One expert reminds us that drone strikes and the arrest of key leaders can be effective against smaller and more traditional terrorist groups, but not against most radicalized and jihadist groups.  Paradoxically, some U.S. allies, such as Pakistan, who often cooperate with Washington, provoke terrorist activities by their very authoritarian policies and practices.  The U.S. needs as many allies as possible in its military counterterrorism efforts, but some of those allies are likely to prove as problematic as drone strikes in the broader effort to prevent and contain terrorism by winning over the hearts and minds of the people.

Mahmood Monshipouri, PhD, University of Georgia, is a professor of international relations at San Francisco State University and he is also a visiting professor at UC-Berkeley, teaching Middle Eastern Politics, and editor, most recently, of Information Politics, Protests, and Human Rights in the Digital Age (New York: Cambridge University Press, 2016).

William V. Dunlap, MPhil, University of Cambridge; JD, Yale University, is a professor of law at the Quinnipiac University School of Law. He teaches constitutional, criminal, national security, counterterrorism, and international law. He is a former associate dean for faculty research and a former associate dean for academic affairs. He has served as chair of the Section on International Law, the Section on Admiralty and Maritime Law, and the Section 3on Internation

This post is based on Paul Rogers’ Monthly Global Security Briefings for Oxford Research Group and was originally posted  on 31 July, 2014. At the time of writing (31 July), Israeli Operation Protective Edge had exceeded the previous major operation, Cast Lead of 2008-9. Both operations have involved intensive use of air strikes combined with major ground incursions. The current war is already longer than the 2008-9 war, with no end in sight. Indeed, by the end of July, positions were hardening and prospects for anything longer than brief further humanitarian pause seemed remote. This briefing provides some context for the conflict together with a preliminary analysis of possible consequences.

The War So Far

Iron Dome in Operation Protective Edge Source: Wikipedia

The current war started on 8 July with intensive Israeli air and artillery assaults on Hamas paramilitary targets, intended primarily to destroy or greatly limit the Hamas ability to fire unguided rockets over much of Israel. In spite of the level of force used, the rocket fire continued, amidst growing concern within the Israeli Defence Forces (IDF) that paramilitaries had constructed many more infiltration tunnels than had been realised. A ground assault followed the initial air assault, with this being intended to destroy rocket launch facilities and stores and also interdict tunnels. As a consequence of this assault, the IDF suffered many casualties, including the deaths of 13 men from the elite Golani Brigade in a single day (20 July). Even after 10 days of conflict, with intensive IDF operations against the infiltration tunnels, Hamas paramilitaries managed to get under the border and in a brief attack killed five young IDF sergeants on a leadership training course. One Hamas paramilitary was killed but others appear to have returned to Gaza. Over the course of the war so far, Israeli forces have struck at over 3,700 targets in Gaza while more than 2,700 rockets have been launched by Hamas and other groups from Gaza towards Israel. The death toll among Palestinians exceeds 1,350 and is rising markedly each day. At least 6,000 people have been injured. Israel has lost 56 soldiers and three civilians, and more than 400 soldiers have been wounded. On 31 July, the 24th day of the war, Israel announced the calling up of a further 16,000 reservists, to bring the total call-up to 86,000. There has been considerable controversy over the numbers of civilian casualties in Gaza, especially the hitting of schools, hospitals and a market. UN figures indicate that at least 70% of those killed are civilians, and nongovernmental international support for Hamas has increased substantially. Public opinion in Israel remains very strongly in favour of continuing the war as a means of stopping the rockets and destroying the infiltration tunnels.

Support for the Adversaries

Hamas: In the past three years, Hamas has lost much of its international support from governments in the region, even though Gaza has existed in what amounts to an open prison controlled by Israel. The Egyptian government of President Sisi is strongly opposed to the Muslim Brotherhood and regards Hamas as a part of this wider movement. The consequent near-total closure of the common border with Gaza and the control of access tunnels has had a marked economic effect on Gaza, exacerbating its siege status. Furthermore, Hamas’s support for Islamist paramilitaries in Syria has lost it the support of the Assad regime in Syria and, to an extent, of the Iranian government. The recent rapprochement between Hamas and Fatah survives, if currently strained, but causes the Israeli government considerable concern. In spite of all the limitations, Hamas’s paramilitary wing has been able to assemble many thousands of rockets and mortar rounds and has also built a network of robust infiltration tunnels that has greatly exceeded Israel’s pre-war estimates. Hamas as a movement retains considerable support in Gaza, with even the impact of the Israeli air and ground assaults having little effect. Israel: Israel retains a measure of support of many western governments but there are growing concerns at the civilian losses in Gaza. The IDF and the defence industry as a whole have very close connections with their US counterparts. The key missile defence system, Iron Dome, is essentially a US-Israeli joint production, including current plans to set up a new production line in the US. Israel is also able to use US munitions stored in Israel. The US is in a position to put very heavy pressure on Israel but is deeply reluctant to do this at present, mainly because of domestic support for Israel. This support remains high but is declining

Cast Lead and Protective Edge

Both the 2008-09 and 2014 Israeli operations have had similar aims – to so damage Hamas that it is massively restricted as a threat to Israeli security. A comparison of the operations so far is indicative. Cast Lead lasted 23 days and ended with a ceasefire brokered largely by Egypt. During that period, Hamas and other groups launched 750 rockets and mortars, all relatively short-range. Israelis lost 13 lives, four of them to friendly fire. Israel killed 1,440 people in Gaza, claiming that half were militants, though Hamas denied this. Since the 2008-09 operation, Israel has deployed the Iron Dome system, and this has intercepted the great majority of approximately 2,700 rockets and mortars fired during Protective Edge.  Hamas has, though, hugely increased its capabilities over the past six years, in spite of its recent political isolation, and has exacted a much higher toll on IDF soldiers during the current ground assault than in 2008-09: 56 so far compared with 13 before. In this sense, the aim of Cast Lead – to substantially degrade Hamas’s crude offensive systems – was a singular failure. Even with the Iron Dome system, vulnerabilities have been demonstrated by the closure of Ben Gurion Airport to several international carriers for several days last week, following a rocket which penetrated the missile shield and landed within a mile of what is Israel’s gateway airport. The loss of so many Israeli soldiers may still seem small compared with the huge losses in Gaza, but the IDF is held in very high regard in Israel.  Indeed, support for the war has likely increased because of these losses and the partial closure of the airport. These appear to have combined to convince many Israelis that, though Hamas is weak and hugely restricted in its location, it represents such a threat to Israel that a protracted war is, if need be, fully justified. The phrase “impregnable in its insecurity” has sometimes been applied to Israel and it is useful in understanding the outlook of a very powerful country that still feels vulnerable.

The home of the Kware’ family, bombed by IDF forces. 8 civilians, including 6 minors, were killed. Gaza, 8 July, 2014. Source: B’Tselem

What Now?

At the time of writing (31 July) it is possible that another humanitarian pause might be agreed and might lead to something more substantial. Assuming that this does now happen, the indications are that the IDF will continue its operations to destroy rockets and tunnels, and Hamas paramilitaries will resist. Given the IDF casualties to date, a pattern is likely to emerge in which urban counter-paramilitary operations will prove both difficult and costly, and the IDF will rely much more on its huge firepower advantage. This is very much what happened with US and coalition forces in Iraq from 2003, and even more so with the Israeli siege of West Beirut in 1982 when at least 10,000 people were killed, the great majority of them civilians. It is already evident that targeting has moved on to the more general Hamas infrastructure, but the very nature of the densely populated Gaza Strip means that the infrastructure for the whole community is also hugely affected. Given the existing impoverishment of the area, the human consequences will be severe, as UN staff have been pointing out repeatedly.

Consequences

In all of its operations against Hamas – Cast Lead in 2008-09, the more limited air assault in 2012, and the current war – Israel has sought to severely damage Hamas’s paramilitary capabilities, and decrease its domestic support. In the first two conflicts that objective was not achieved, and it is unlikely that Israel’s current operation will succeed this time around. In spite of Hamas’s greater international isolation, its paramilitaries have this time had a substantial impact on the IDF, and the movement retains domestic support. Moreover, international public opinion has moved heavily against Israel. One of the major changes in comparing the current war with the two previous wars is that the use of social media has hugely expanded, resulting in graphic images being distributed across the region and beyond in near-real time. One effect of this, in turn, is that the more conventional western media reporting is itself becoming more graphic. In spite of a very efficient Israeli information operation, this change is working against Israel’s interests. It also means that Islamist propagandists across the Middle East and beyond are easily able to present the war as a further example of “Zionist aggression”. Indeed, they will also relentlessly point to close US-Israel links, further developing their long-term image of a “Crusader-Zionist war on Islam”, in spite of Secretary of State Kerry’s undoubted personal commitment to achieving a ceasefire. The long-term consequences of this are difficult to read, but could give a boost to radicalisation well beyond Israel and the occupied territories. That alone is an added reason why a ceasefire at the earliest opportunity is not only desirable but essential.

Many have argued that civil wars are more likely to occur along religious divisions. But evidence indicates that intrastate conflict is actually more likely within linguistic dyads than among religious ones.

In the 1990s Samuel Huntington argued that conflict across civilizational or religious lines would replace the ideological divisions that had defined political struggles during the Cold War period. Opining that Islam has ‘bloody borders’, he believed that conflicts would be particularly prevalent between ‘Muslims’ and ‘non-Muslims’. This led Huntington to further suggest that a future clash between ‘Islamic civilization’ and the West might occur.

Since September 11^th 2001 and the subsequent proclamation of the “War on Terror,” Huntington’s thesis has gained widespread attention among political leaders and citizens around the world. In 2014, for example, Tony Blair asserted that “religious difference will fuel this century’s battles.” During the 2016 US presidential campaign, President-elect Donald Trump seemingly subscribed to Huntington’s ideas when calling for “a total and complete shutdown of Muslims” entering the country to prevent violent attacks on US citizens.

Whereas many social scientists would agree that ideological conflict between communism and capitalism, both between and within states, has declined since the end of the Cold War, no agreement exists about what, if anything, replaced ideology. Most scholars who study internal conflict or civil war would not distinguish between linguistic, religious, and racial markers but rather classify these categories as part of the larger concept of ethnicity. Yet some conflict researchers follow Huntington and identify religious differences as particularly conflict-prone. In doing so, important alternatives such as ethno-nationalist mobilization based on linguistic identities often receive too little attention.

Are internal conflicts mostly about religion or language?

Image by Christiaan Triebert/Flickr.

In a study that is forthcoming in the Journal of Conflict Resolution, I and my co-authors Lars-Erik Cederman (ETH Zürich) and Manuel Vogt (Princeton University) conduct such a comparison. We analyse the probability of internal armed conflict between linguistically and religiously distinctive groups between 1946 and 2009.

Contrary to Huntington’s thesis, linguistic differences show a strong and robust relationship with the outbreak of intrastate conflicts. In fact, we find that linguistic divisions are more conflict-prone than religious differences.

These findings continue to hold when we focus only on the years since 1990 – the period to which Huntington’s thesis should be the most relevant. Our results further suggest that in no world region are religious differences more likely to be associated with internal armed conflict than linguistic divisions. We find the strongest support for a greater conflict-proneness of language compared to religion in Eastern Europe and Asia.

Even in the Middle East, we find a slightly higher, if uncertain, probability of armed conflict across linguistic than religious lines. The Middle Eastern finding at least in part results from multiple Kurdish rebellions in Turkey, Iraq, and more recently Syria as well as smaller uprisings of linguistic minorities in Iran.

When focusing only on conflicts that involve Muslim groups, we do not find substantial differences to other world religions. Although the majority of all armed internal conflicts today take place within Muslim-majority states, the majority of Muslim groups do not engage in violent rebellion. Our analyses also reject the thesis that Muslim groups disproportionately engage in conflict with non-Muslim groups.

Why Linguistic Differences?

Instead linguistic differences continue to be more frequently related to armed uprisings within states. While the brutal civil war in Syria captured headlines over the past years in many Western countries, destructive conflicts across linguistic lines haunt South Sudan, Burma, and Turkey.

Of course, linguistic differences are more widespread than religious divisions. In other words, ethnic groups in any given country are more likely to be divided by language than by religion. Notwithstanding these differences in frequency, our results indicate that linguistic divisions are disproportionately more often related to armed conflict than religious distinctions.

In our forthcoming article, we argue that it is the power of nationalism that makes linguistic divisions more conflict-prone than religious ones. Language gained political relevance in the late 18^th century when the French Revolution transferred political authority from absolutist rulers to the people. About the same time that political power became vested in European peoples, the industrial revolution created incentives to further homogenize European nation-states. Mass schooling and mass newspapers laid the basis for imagined national communities.

These developments provided both motive and opportunity for violent conflict across linguistic boundaries. Where members of ethnic groups are barred from having their children taught in their native language or experience linguistic discrimination in the job market and their interaction with the state, some of them will voluntarily assimilate into the dominant culture, but others develop grievances and may even refuse assimilation.

The elites of such discriminated groups can voice these grievances through publications in their own language and use it to express nationalist aspirations and demands. When the host state is unable or unwilling to address these demands, violent conflict becomes more likely. These dynamics are illustrated by Sri Lanka’s decades-long civil war between Singhalese and Tamils, and recurrent Kurdish rebellions in Turkey.

Given the link between industrial advancement and language-based nationalism it is unsurprising that we find higher rates of linguistic conflict in the relatively highly developed regions of Eastern Europe and Asia rather than in Sub-Saharan Africa. Central and Eastern Europe may even be considered as the cradle of linguistically-based nationalism.

The multi-ethnic Austro-Hungarian and Ottoman empires did not fulfil the modern creed of “one people, one state,” and violently disintegrated during World War I. The Soviet Union and Yugoslavia followed suit in the early 1990s. To this day, Turkey has not come to terms with its Kurdish minority, and once more experiences internal conflict.

Yet the idea of nationalism did not remain contained to Europe. A highly flexible concept, it informed the national liberation struggles of former colonial subjects against the European colonial powers. The lines of division here were usually race and language rather than religion. For decades, the Israeli-Palestinian conflict was a conflict between Hebrew-speakers of European origin and Arab speakers, who had lived in Palestine for centuries. Only in the past two decades has it taken on more religious undertones.

Policy Solutions?

Linguistic and religious differences will remain with us for the foreseeable future. However, very few of these fault lines can be expected to erupt in violent conflict. Whether linguistic differences transform into seemingly incompatible nationalist projects, or whether religious divisions into ostensibly intractable positions, depends on how political leaders from different groups interact with one another.

Frequently armed rebellion emerges in politically highly exclusive and discriminatory contexts. Where political leaders with specific linguistic or religious backgrounds are barred from decision-making that affects their groups, conflict is more likely to break out than in states where they have some influence in government circles. Exclusion along ethnic lines creates clear insiders and outsiders, fosters grievances among the excluded, and suggests that there is “no other way out” but violent resistance. Zimbabwe, both under Smith and Mugabe, Iraq under Saddam Hussein, and Burma to the present day are examples of ethnically exclusionary regimes. Each of these states also experienced violent rebellion by excluded groups.

In another joint study published in “Peace and Conflict 2016”, we show that excluding political elites with different linguistic or religious backgrounds from governmental power is pervasive in the Middle East and North Africa. So is political discrimination that denies the Palestinians in Gaza citizenship rights, keeps the Shia from voting in Qatar, or persecutes Kurds for political reasons in Turkey.

Figure 1 displays the average population share that experiences discrimination for different world regions and years. The data derives from the Ethnic Power Relations (EPR) dataset available at https://growup.ethz.ch/pfe.

Figure 1 reveals that ethnic discrimination remains staggeringly high in the Middle East although the region has experienced some improvements over the past twenty years. That religious differences in the Middle East erupt into violent conflict may be less of a surprise once this context is taken into account.

Our research thus suggests that avoiding ethnic exclusion and discrimination through power-sharing in multi-ethnic governmental coalitions will reduce the likelihood of armed conflict across both linguistic and religious lines. Elite accommodation in power-sharing coalitions has contributed to greater stability in such diverse places as Bosnia, Nigeria, Burundi, and Malaysia regardless of the type of ethnic differences. Although no panacea, power-sharing is associated with a substantial decrease in the likelihood of internal armed conflicts compared to exclusive environments.

While there has been a trend towards ethnic accommodation since the end of the Cold War, we do not know enough about its origin. Future research needs to investigate the causes of accommodation in greater detail and pay particular attention to appropriate solutions for violent conflict across linguistic lines relative to religious differences.

Nils-Christian Bormann is lecturer and Humanities and Social Science Fellow in the Politics Department at the University of Exeter.

Manuel Vogt is a visiting postdoctoral research associate at Princeton University and senior researcher at ETH Zürich.

Lars-Erik Cederman is Professor of International Conflict Research at ETH Zürich and the author of Inequality, Grievances, and Civil War (Cambridge University Press, 2013).