A recent article by Scott D. Sagan of Stanford University and Benjamin A Valentino of Dartmouth College in International Security (Sumer 2017) in conjunction with Bertrand de Jouvenel’s 1945 book Du Pouvoir (On Power for the English translation) help us think about nuclear weapons. One characteristic of these weapons is that they are designed to be used against civilian populations (although there also exist “tactical,” nuclear weapons, besides “strategic” ones, capable of use against advancing armies or military installations). “If you kill my people,” Leviathan seems to proclaim, “I will kill yours too.”
Bertrand de Jouvenel, a French political philosopher, wrote On Powerfrom his refuge in Switzerland during World War II. He observed how both the German government and the Allies had resorted to bombing civilians. The book was written before the nuclear attacks on Hiroshima and Nagasaki, which provided even more terrible illustrations. While aristocratic governments generally kept ordinary people out of their conflicts, Jouvenel claimed, “totalitarian democracy” has regimented ordinary citizens into the states’ wars; they have become implicit combatants. He wrote:
In the time of Louis XIV … conscription was unknown, and the private person lived outside the battle … For the first time in [American] history, a President of the United States [Franklin D. Roosevelt] looked on the mass of his fellow-citizens as ‘human potential,’ to be used as might best serve the prosecution of the war! … Whereas the feudal monarchs could nourish hostilities only with the resources of their own domains, their successors have at their disposal the entire national income.
Sagan and Valentino vindicate Jouvenel by arguing that American public opinion favors the use of nuclear weapons. According to pollsters in August 1945, 85% of Americans approved of the bombs just dropped on Hiroshima and Nagasaki. Today, in retrospect, less than 50% think it was a good idea. But Sagan and Valentino’s 2015 opinion survey indicates that, in a similar scenario with Iran instead of Japan, 59% of Americans would support a US government decision to nuke 2 million civilians in order to prevent the deaths of 20,000 American soldiers in an alternative ground invasion. (The scenario presented to respondents was that of a war started by the Iranian government.) And most of these hawkish respondents would not change their minds even if a diplomatic solution were possible.
Of the American respondents who favored either a nuclear or conventional strike on civilians to save American soldiers, 68% agreed with the statement: “Because the Iranian civilians described in the story did not rise up and overthrow the government of Iran, they must bear some responsibility for the civilian fatalities caused by the U.S. strike.” Sagan and Valentino express surprise at the number of respondents who “suggested that Iranian civilians were somehow culpable or were less than human.”
This reading of American opinion contradicts Thomas Schelling who, in his 2005 Nobel lecture, argued that a strong convention had developed against the use of nuclear weapons. It also contradicts the principle of “noncombatant immunity”: as Sagan and Valentino write,
Both just war doctrine and the laws of armed conflict require leaders and soldiers to make active efforts and accept risks in war to avoid the deaths of foreign civilians.
We cannot discount an explanation in terms of nationalism as a modern version of tribalism. But a related and perhaps deeper issue is the idea that the state and the citizens form an undivided whole. And here, we meet what may seem like a paradox. The more democratic the state is, the smaller the difference between its rulers, combatants, and civilians, and the more justifiable should be a deliberate attack on the latter, ceteris paribus. This is especially true if democracy refers not to majoritarian democracy but is based on a sort of unanimous (implicit) social contract where the state is, at some level, identical to its citizens. The more dictatorial the state, on the contrary, the larger is the distance between rulers, combatants, and civilians, and the less justifiable should be a deliberate attack on the latter, even in a just, defensive war.
The first lesson of these reflections seems to be that, under any reasonable moral standard, using nuclear weapons against the civilian populations of dictatorial regimes – like Iranians or North Koreans – must be morally unjustifiable, even in a just war against their Leviathan. Disagreeing with this, as many Americans appear to, is like legitimizing the state as a mass killer. (We don’t have comparative data on public opinion elsewhere in the world.) This suggests that de Jouvenel’s concern remains very relevant.
These moral considerations seem far removed from the positive economics of war. But welfare economics has taught us that any public policy recommendation ultimately depends on moral judgements because it harms some individuals while it favors others.
In a war between two democratic countries, it would be inconsistent to advocate both the civilian immunity principle and the idea (or the fiction) that the democratic state represents all its citizens. But perhaps democratic states are less likely to engage in wars? The case of Switzerland springs to mind, but it seems contradicted by the American example. Of course, a sample of two does not have much explanatory power.
The second lesson is that we should ponder the question of whether the state should not be conceived as a Nozickian, arms-length protection agency, instead of an association of all its citizens – even a constitutional association à la Buchanan.
None of what I have said argues against a defensive war (or preemptive attack) against a foreign tyrant, but it does impose tight limits on the means used and on the cost imposed on foreign civilians. General Paul Selva recently put it in neat terms before a Senate committee: “We take our values to war” – assuming of course that these values are independently defendable, as I believe classical-liberal or libertarian values are. This brings us back to a simple but forceful idea, close to de Jouvenel’s thesis: if “we” use liberticidal means to defend liberty, there will ultimately be no liberty left to defend.
Pierre Lemieux is an economist affiliated with the Department of Management Sciences of the Université du Québec en Outaouais.