Testing the Waters: How Common is State-to-State Conflict Over Water?

Testing the Waters: How Common is State-to-State Conflict Over Water?

August 7, 2008 By Rachel Weisshaar

I was lucky enough recently to vacation in Israel—and I still have the jet lag to prove it. On the second day of the trip, as we crossed the Jordan River and entered the Golan Heights, our guide explained that people have fought over water throughout history—especially in the Middle East. “Aha!” I thought to myself. “Another example of how the average person mistakenly believes that water scarcity leads to conflict—whereas, as an Environmental Change and Security Program staff member, I know that interstate ‘water wars’ are actually incredibly rare.”

Yet our guide proceeded to describe several water-related conflicts between Israel and its neighbors before and after the 1967 Arab-Israeli War. So when I returned to the States, I was inspired to look up these events in the Pacific Institute’s Water Conflict Chronology and in Oregon State University’s Transboundary Freshwater Dispute Database. I found that Israel and its neighbors were frequently engaged in violent conflict with one another over water during the 1960s and early 1970s. For instance, after Syria began diverting the headwaters of the Jordan River in 1965—a project that would have deprived Israel’s National Water Carrier of approximately 35 percent of its water and the country as a whole of around 11 percent of its water—Israel responded with a series of military strikes against the diversion works. The back-and-forth attacks helped instigate the 1967 Arab-Israeli War.

Another example: In 1969, Israel, which believed that Jordan was overdiverting the Yarmouk River, bombed Jordan’s East Ghor Canal. The United States mediated secret negotiations in 1969-1970, and the Jordanians were allowed to repair the canal in exchange for abiding by Johnston Plan water quotas and expelling the Palestinian Liberation Organization from Jordan.

So how do we explain the apparent disconnect between the numerous instances of violent conflict over water in the Middle East and political scientists’ insistence that water rarely leads to interstate conflict? I think the answer is twofold. First, I happened to be standing in the region of the world that is by far the most prone to conflict over water. There were only 37 violent interactions over water between 1946 and 1999, and 30 of these were between Israel and a neighbor. Water experts recognize that the Middle East is the exception to the general pattern of water disputes leading to cooperation, not conflict. But to Middle Easterners like my Israeli guide, it may indeed seem that water frequently leads to conflict.

Second, the devil is in the definitions. The Transboundary Freshwater Dispute Database classifies events according to its Water Event Intensity Scale, which runs from -7 (“formal declaration of war”) to 7 (“voluntary unification into one nation”). The East Ghor incident is classified as -6 (“extensive war acts causing deaths, dislocation or high strategic cost”). Events classified as -6 can include “use of nuclear weapons; full scale air, naval, or land battles; invasion of territory; occupation of territory; massive bombing of civilian areas; capturing of soldiers in battle; large scale bombing of military installations; [and] chemical or biological warfare,” so they seem to differ from war only in that war has not been formally declared.

Perhaps this is where much of the confusion comes from: Political scientists studying water conflict use a very narrow definition of war—probably a lot narrower than that of most non-experts. So while the average interested citizen would likely call the 1965-1967 conflict between Israel and Syria over the National Water Carrier and the Headwater Diversion project a war—or at least a high-level interstate conflict—the political scientist studying water conflict and cooperation would not.

Now, as a general principle, I’m all in favor of precise language and definitions. But formal declarations of war seem to have gone out of fashion over the past half century; the United States, for instance, has not formally declared war against another country since World War II. If the current war in Afghanistan were over water—which it decidedly is not—would it still merit only a -6 on the Water Event Intensity Scale because the United States has not formally declared war against the Taliban? It seems that requiring a formal declaration of war to classify a conflict as a war is perhaps defining the term too narrowly.

But although political scientists may be to blame for clinging to a somewhat outdated definition of war, the media are perhaps at fault for using the word too broadly in an attempt to make their headlines more enticing. This editor concludes—only somewhat self-servingly—that we would all benefit from using language more precisely. I welcome your responses.

Photo: The Golan Heights landscape still bears scars from the 1967 Arab-Israeli War. Courtesy of Rachel Weisshaar.

Topics: conflict, security, water
- Tom Deligiannis
  
  Thanks for this interesting post. I think that it raises an important issue in environment conflict research beyond conflict or cooperation over water resources. The issue of how war or conflict is defined in environment conflict research has varied depending upon which study you examine, and those definitions have had an important part to play in how that research has been received or rejected by scholars and policy makers.
  
  Homer-Dixon started his research into environment-conflict linkages by looking at war in the traditional manner of security scholars in the early post Cold War era – wars between states or wars that would overturn or undermine state authority. His early findings ruled out inter-state war as a result of environmental scarcities, and he focused most of his energy in later projects on sub-national wars. But even these scenarios were the type of conflicts normally studied by security scholars – regime changing or regime threatening sub-national conflicts – like the Chiapas rebellion, Rwanda genocide, South Africa case, etc. Colin Kahl has correctly noted how the state-centred focus of Homer-Dixon and Jack Goldstone’s work (from which Homer-Dixon drew a great deal) was an essential feature of the 1990s environment conflict projects.
  
  Gunter Baechler’s studies were more expansive in their definition of the dependent variable – conflict. The definition of conflict used in the Swiss group’s case studies of environment-conflict links ranged from interstate violent conflict, interstate non-violent conflict (or what some might call political disputes of different kinds), to sub-national conflicts ranging from regime changing events to small-scale local conflicts.
  
  Baechler’s focus on local environmental violent conflicts, however, hasn’t been followed up by scholars as much as those focusing on sub-national regime changing/threatening conflict or international conflict. Colin Kahl’s work was in the tradition of Homer-Dixon with its focus on violent conflict directed by the state or elites – violence that is still quite extensive or intense within the state, such as his Kenya case, or insurgency conflicts in his Philippine case (The state exploitation violence he describes is often made up of many local conflicts, however.) Ohlsson’s work brought the issue of livelihood conflicts to the research agenda, but he didn’t pursue it very far, and his detailed case study of Rwanda was still focusing on rather large scale intra-state violence.
  
  The lootable conflict resource body of research – Collier, Ross, etc. – defines the conflict variable widely, but one detects a greater focus on large-scale inter or intra-state violence in much of this work. Lootable resources are often seen to be fueling large scale sub-national insurgencies, insurrections, secessionist movements, or ethnic/political conflicts. This bias is reinforced by the fact that the statistical data often used by the lootable crowd often draws on data bases where violent conflicts were only recorded if more than 1000 battle deaths resulted. Many of PRIO’s early studies into environment-conflict linkages similarly used this 1000-death violent conflict threshold, and their failure to find many examples of environment-conflict correlations that exceeded this threshold fueled much of the skepticism of environment scarcity-conflict links coming out of PRIO in the late 1990s and early part of this decade.
  
  PRIO has come up with a new conflict database, however, that lowers the threshold to 25 deaths, and recent research by scholars such as Urdal is now much less skeptical about environment scarcity-conflict linkages.
  
  My own research in Peru has convinced me that the definition of conflict and war has been a major hindrance to the recognition and appreciation of environment scarcity -conflict links. I found many local conflicts in Peru (and other parts of Latin America) where small numbers of people were killed or injured. In some cases, as in Peru, these disputes also fed into larger state-wide insurgencies to both worsen the general violence of the insurgency, or give it a particular flavor or pattern. Looking around, I see many other examples of this type of violent conflict over scarce or abundant environmental resources in different parts the world. As Homer-Dixon concluded, this violence is often sub-national and diffuse. So, it’s going to escape the gaze of those only looking for inter-state events, or major regime changing, or secessionist conflicts and insurgencies.
  
  While the new PRIO database is a good step toward systematically recording these conflicts, we need to supplement and improve this database through more local, on-the-ground studies. Many of the conflicts that I came across in Peru were not widely known about outside of a small group of area experts, and they rarely if ever were publicized outside the country in the newspapers and reports used by those compiling conflict databases. This is perhaps a historical artifact back to the days before the 24/7 global news cycle, since my work was focused on the period before 1980. It’s easier now to get reports on your desktop on local conflicts from far flung news agencies and reporting services, especially if you know the local language. It’s even better if you can also visit the areas and compile reports from the variety of local sources and documents available, as Rachel’s post nicely demonstrates.
  
  Numerous environmental security scholars have called for more local studies, but few are produced and few large-scale projects to commission local studies have resulted. I hope that soon changes.
  
  Tom Deligiannis
- http://www.blogger.com/profile/10500451930199560177 Juan Dumas
  
  I may be coming in a bit late in commenting this post but I would like to suggest another reason for more local conflict studies.
  
  Our belief at Futuro Latinoamericano is that environmental conflicts at the local level in Latin America are the expression of inconsistencies between national public policies.
  
  While official documents can state an “explicit” environmental policy, the real and “implicit” environmental policy is evidenced in the impacts of other policies – trade, investment, agriculture, health, transportation, industry- that do not adequately integrate environmental concerns in their formulation processes.
  
  Carefully documenting environmental local conflicts and aggregating this information at a national level can show how failure to integrate cross-cutting environmental factors in other policy arenas can result in serious threats to national governance.
  
  We are trying to raise funds for regional research on this.

New Security Beat

NewSecurityBeat

Testing the Waters: How Common is State-to-State Conflict Over Water?

Join the Conversation

Featured Media

New Security Beat

NewSecurityBeat

Testing the Waters: How Common is State-to-State Conflict Over Water?

Join the Conversation

Trending Stories

Featured Media

Related Stories