When (In)Action Speaks Louder Than Words: Confronting the Collapse of Humanitarian Values in Foreign
“Why do good people and their governments repeatedly turn away from intervention that could halt genocides and other mass abuses of human beings?”
“What devaluation of human lives could possibly allow this?”
I began to examine such questions when I became aware of the indifference toward the vast scale of atrocities being perpetrated in Darfur, Sudan. I saw a connection between earlier research I had published with David Fetherstonhaugh and colleagues in 19971 and subsequent research with Deborah Small and George Loewenstein.2 Specifically, this work documented the insensitivity to large numbers of lives at risk that we labeled psychophysical numbing, consistent with the general nonlinear model of valuation proposed by Daniel Kahneman and Amos Tversky (1979) in their landmark paper on prospect theory.3
Subsequent studies uncovered additional evidence of insensitivity described as compassion fade and, in some cases, compassion collapse, where valuation actually decreases and may even collapse to zero as the number of lives at stake increases.4 All of this helps explain why many who care greatly about individual lives lose their enthusiasm and compassion when the numbers get large. Slovic, Zionts, Woods, Goodman, and Jinks proposed some procedures that might infuse the dry statistics of atrocities with enough emotion to motivate people to care enough to act.5
But yet another problem beyond numbing, at the very foundation of valuation, decision-making, and action has become apparent. It is easy to view inaction on the part of powerful and well-intentioned governments as resulting primarily from a lack of direction or pressure from a numbed public that places little value on saving foreign lives. In her Pulitzer Prize winning book, Samantha Power concluded that: “[G]enocide in distant lands has not captivated senators, congressional caucuses . . . lobbyists . . . or individual citizens. . . . The battle to stop genocide has thus been repeatedly lost in the realm of domestic politics. . . . It takes political pressure to put genocide on the map in Washington.”6
True enough, and a testimony to the indifference on the part of a numbed citizenry that doesn’t feel an emotional connection to different suffering. But genocides and mass atrocities of recent years in places such as Rwanda, Darfur, and Syria have been thoroughly documented while they were taking place and that documentation has included emotionally jarring testimony and images. Something more than numbness must be impeding action. Is it a matter of values?
It was easy to blame the Bush administration for being uncaring. Even when Colin Powell returned from Darfur in 2004 and used the “G word” to describe the atrocities there, no action was taken.7 Then the Obama administration came to power and hopes soared. Surely President Obama cared about humanitarian causes, and he was surrounded aides, including Samantha Power, who certainly placed high value on human lives. In fact, Ambassador Power has recalled that at her first meeting with the President he was greatly interested in her book and its documentation of repeated failures of the American government to live up to its ideals.8 Yet six years into an administration most likely to succeed in taking action to mitigate or halt mass atrocities, little seems to have changed—certainly not in Darfur. What’s going on?
Perhaps psychology can again provide some explanation. Aided by valuable discussions with colleagues Robin Gregory, David Frank, and Daniel Västfjäll, I sought insight from data I began collecting in 1961. That research examined how people made decisions between two options that were equally valuable to them.9 Consider a gift package made up of two components—cash and a coupon worth $X, redeemable at a store you like. There are two such packages. Package A gives you more cash than Package B, but you are able to increase B’s advantage in the value of the coupon book by enough to make the two gifts equally attractive to you. But you have to choose. Which one would you take? Naïve theories would predict you will flip a coin, thus being equally likely to choose A or B. That didn’t happen. Eighty-eight percent of respondents, each of whom had individually adjusted the packages to make them equally attractive for them, made the choice in the same direction.10
What do you think they did? If you predicted that they chose the package with the greater amount of cash, you are correct.11
Similar results occurred with choices among nine other pairs of two-attribute “bundles,” such as baseball players (described by batting average and number of home-run hits), secretarial applicants (described by typing speed and typing accuracy), and so on.12 In every case, choices among individually equated pairs were highly predictable, with about eighty percent of respondents adhering to the following the rule: choose the option that is better on the attribute that is inherently more important (e.g., cash, or batting average, or typing accuracy).13
Not trusting the adjustment method used to equate the options in each pair, I finally found a way to overcome my methodological concerns and published these findings after more than a decade.14 Some thirteen years later, Amos Tversky and Shmuel Sattath incorporated this “more important dimension effect” into a new theory of choice.15 The core finding was named “the prominence effect.” 16 The essence of this effect is that, although we may have a qualitative sense of the importance of valued attributes, we may not have a sense of the appropriate quantitative tradeoffs when these attributes compete with one another.17 For example, we highly value both affordability and safety in a car, but how much more we should be willing to spend for a specific increment in safety is by no means obvious to us. We struggle with making tradeoffs and seek a simple, defensible way to choose among options whose attributes are important but conflicting. Here is where the prominence effect enters: don’t struggle to perform any quantitative calculations to weigh and compare valued objectives. Choose what is best according to the most prominent—that is the most defensible—attributes. You can’t go wrong. You can well defend your choice to yourself and others. Moreover, it likely “feels” right.
So what does this have to do with valuing foreign lives and genocide? A lot.
It is well recognized that decisions to save civilian lives by intervening in foreign countries’ domestic affairs are among the most difficult and controversial choices facing national decision makers. Although each situation is unique, such decisions typically involve tradeoffs that pit the value of human lives against other important objectives. And on rare occasions we do decide to intervene. In 2011, the United States supported military action to protect the lives of civilians living in Libya,18 and more recently the American military intervened aggressively to protect a threatened population of Yezidi people in Iraq.19 On the other hand, the United States has done little to intervene in the genocide in Darfur or to halt the barrel bombings and other government-led atrocities in Syria that have led to hundreds of thousands of deaths and millions of displaced people.20
The inconsistencies are striking. Why intervene in some situations and not in others that, by the numbers, seem far worse? And how do we reconcile the immense value our society places on an individual life with our failure to respond to the plight of millions? What are our true values when it comes to saving human lives? Should we accept these inconsistencies? Are we oblivious to them? Are other objectives really important enough to outweigh millions of lives? Why can’t sheer numbers, once great enough, tip the scales toward at least some forms of meaningful intervention if not outright troops on the ground?
The lightbulb switched on in 2012 when I was at a conference in Jerusalem, listening to a presentation by social psychologist Nurit Schnabel, and her colleagues Ilanit SimanTov-Nachlieli and Arie Nadler.21 The title of their talk says it all: “Sensitivity to Moral Threats Increases When Safety Needs Are Satisfied: Evidence of Hierarchical Organization of Psychological Needs.”22 Their conclusion jumped out at me: consistent with Maslow’s hierarchy of needs, the basic need for security must be satisfied before people will respond to higher order needs.23 In other words, when security is tenuous moral action to help others is unlikely.
Immediately I thought of the prominence effect, where determiners of choice are found to be hierarchical (we called it lexicographical), which means satisfying the most important attributes with little or no compensation allowed for lesser attributes. That is, prominent attributes trump less prominent factors. Prominence is driven by the need to justify and defend one’s decisions and actions in a way not called for when simply stating one’s values. And what is the most prominent set of values in today’s foreign policy world? National security interests! Could the unquestioned importance of national security explain, at least in part, the disconnect between the lofty expressed values of our government for protecting distant lives and the minimal valuation of those lives revealed by government inaction when millions are threatened?
The rhetoric of the two most recent American presidents leaves no doubt that, in terms of expressed values, national security and humanitarian lifesaving are both vital objectives. Speaking on CNN in January of 2009, George W. Bush remarked: “The most important job I have had—and the most important job the next president is going to have—is to protect the American public from another [terrorist] attack.”24
Barack Obama echoed this sentiment, saying: “I have a solemn duty and responsibility to keep the American people safe. That’s my most important obligation as President and Commander-in-Chief.” 25 Yet both leaders expressed equally high valuation for life saving. Speaking at the Holocaust Museum in Washington, DC, President Obama stated that we need to do everything we can to prevent and end atrocities: “I made it clear that ‘preventing mass atrocities and genocide is a core national security interest and a core moral responsibility of the United States of America.’”26
I am drawn to the hypothesis that, because of the prominence effect, lofty humanitarian values are systematically devalued in the decision-making process. When intervention to protect thousands of nameless, faceless lives in a distant land is seen to increase risks to national security, security invariably wins. Decisions in support of security appear vastly more defensible than decisions to protect distant lives. In the few recent situations where the United States has intervened with the stated objective of saving lives, there were presumed security benefits as well, thus no conflict between objectives. We attacked Saddam Hussein, not because he was a mass murderer, but because we believed he possessed and might use weapons of mass destruction.27 When Libyan leader Muammar Gaddafi threatened to go door to door in Benghazi, killing anyone who opposed his regime, the United States joined a NATO coalition to topple him from power.28
But he, too, was considered a threat to security, long seen as a loose cannon addicted to violence at home and elsewhere. His menacing visage adorned the cover of Time magazine four times since 1986, when Ronald Reagan referred to him as “this mad dog of the Middle East.”29 When the United States recently decided to come to the rescue of thousands of Yazidis threatened by ISIS in Iraq, we were also protecting American military and diplomatic personnel stationed in nearby Erbil.30 Without that security objective, would we have aided the Yazidis?
In contrast, humanitarian intervention in Darfur appears to have been blocked by security objectives in addition to the military and domestic political costs. We have long sought to obtain intelligence regarding terrorist operations from the Sudanese government.31 In addition, Sudanese President Omar al-Bashir, who takes a back seat to no one as a murderer, has been protected by the Chinese government, which for many years was the major buyer of Sudanese oil.32 An action against al-Bashir that strained relations with China would have jeopardized U.S. economic and military interests.
One of the most stunning conclusions by Samantha Power in her book The Problem from Hell was that America’s repeated refusals to end genocide were not “accidental products of neglect” but rather “They were concrete choices made by the country’s most influential decisionmakers after unspoken and explicit weighing of costs and benefits.”33
But if the prominence effect is indeed infiltrating top-level policy decisions and causing decision makers to systematically devalue humanitarian actions, I doubt that the decision makers are consciously aware of this. The prominence mechanism driving the decision-making process is not consciously expressed devaluation of distant lives; this would be abhorrent to leaders who truly do value those lives. Rather, I believe that prominent objectives, in particular those offering enhanced security, draw attention away from less prominent goals. All eyes are on options that protect the homeland, and decision makers fixated on the security objectives likely fail to consider seriously the millions of people under siege and left to die. Compensatory weighing of costs and benefits associated with seeking security and saving distant lives is not really occurring.
Thus meaningful action to prevent genocide and mass atrocities faces two psychological obstacles. The prominence effect leads to decisions that favor inaction, even when this contravenes deeply held values. And decision makers can get away with this because the public is psychologically numbed. As Samantha Power observed: “No U.S. president has ever made genocide prevention a priority, and no U.S. president has ever suffered politically for his indifference to its occurrence. It is thus no coincidence that genocide rages on.”34
My colleagues and I have been working to design laboratory experiments to test these speculations about the psychological prominence of security in values revealed through decisions. In a recent pilot study, we posed the humanitarian crisis in Syria (prior to the involvement of ISIS) to respondents instructed to play the role of the U.S. President. The objective of protecting 100,000 civilian lives by creating a “safe zone” was pitted against the decision to not intervene in order to minimize the political and military risks of intervention. We assumed the latter objectives would be prominent. Preliminary results support the hypothesis that an individual’s strongly expressed values for intervening to protect lives are often contravened by that same person’s decisions in favor of nonintervention for the sake of security.
But amidst this sobering view contrasting our stated values and our revealed values, the pilot study did offer a ray of hope that needs to be pursued. We found a strong order effect in our data. One group of our respondents was first asked to think about their values for the competing objectives and to quantify them on a 0–100 rating scale of importance. One of the rated objectives was not intervening in order to protect national interests and national security. The objective of the alternative action, intervention, was characterized by three subgoals:
(1) Create a no-fly zone to stop Syrian airplanes from attacking citizens; (2) Place U.S. troops on the border of the safe zone to protect it; (3) Have U.S. troops accompany ground shipments of food and medicine to ensure their delivery to the safe zone.