The Washington PostDemocracy Dies in Darkness

The Espionage Act has become dangerous because we forgot its intention

The Julian Assange case exposes how changing concepts unintentionally broadened a law

John Shipton attends a news conference at the British Consulate in New York, June 17, 2022, in response to the decision to extradite his son Julian Assange to the United States. (REUTERS/David Dee Delgado)
Placeholder while article actions load

This piece has been updated to clarify that only four of the 17 charges against Julian Assange stem from the leaking of diplomatic cables.

To guard government secrets, the Justice Department has increasingly weaponized the Espionage Act of 1917 — a law that critics have condemned as “overbroad” and “vague.” The department’s 2019 decision to charge Julian Assange, the WikiLeaks founder, with violating the Espionage Act crossed a new line. On Friday, the British government ordered the extradition of Assange on this indictment, although he pledged to appeal. Many observers are expressing concerns about the future of journalism if the government is heavy-handed in applying the Espionage Act.

But originally, the secrecy provisions of the Espionage Act were not the draconian behemoth that people today imagine them to be. In 1917, there was nothing particularly broad nor vague about these provisions. A century later, however, they have become very dangerously misunderstood — and misapplied.

The great danger of these secrecy provisions hinges on a key phrase: The statute protects all information “connected with the national defense.” The act says nothing about what this phrase might mean. Despite how vague that sounds, the Supreme Court declared the law constitutional in 1941. Although the concept of “classified material” did not exist in 1917, everyone now assumes that any secret that is “classified” must be protected by the Espionage Act.

This assumption is wrong. The act has become so dangerous only because our society has entirely forgotten the concept of “the national defense.” Americans in 1917 understood the term intuitively. A few decades later, that understanding started to vanish as Americans embraced a newer, far more expansive concept: “national security.”

During World War I, a lengthy German espionage and sabotage campaign exposed the need for U.S. anti-espionage legislation. As early as December 1915, President Woodrow Wilson fulminated to Congress about the Germans who had “formed plots,” “entered into conspiracies” and “sought to pry into every confidential transaction of the Government.” The following year, German saboteurs blew up an enormous munitions cache in New York Harbor, setting off a deafening explosion that woke people across New York and New Jersey and even damaged the Statue of Liberty. Somehow only six people died.

With U.S. troops headed into harm’s way in 1917, the stakes rose. Congress took up the president’s demand for a new espionage law. A blistering debate raged over a provision to impose censorship on war news, but once this was taken out, the rest of the Espionage Act passed with little controversy, about two months after the U.S. entry into the war. Even the New York Times declared that when “more important to the enemy than to our own people,” it was “the business of the government to keep secret military facts.”

Designed to keep military facts a secret — to protect the Army and Navy as the United States began to fight a brutal war — Congress had no need in 1917 to define “the national defense,” because of the consensus understanding of the term. In early 20th-century America, the concept of the national defense was ubiquitous. People thought of America’s international position in terms of its “national defense.” They debated the country’s “national defense” in newspapers, journals and Congress. The House floor manager of the Espionage Act was completely unworried about the term: National defense’s “meaning is pretty well understood in the minds of the public,” he argued.

“National defense” was understood narrowly, but with a sense of elasticity in case of war. It encompassed the military and its needs, and agriculture. Agriculture was central to defense because a war could be lost as easily through civilian starvation as through battlefield defeat. Then, if the military needed them to fight a war, vast sectors of the American economy — including production, transportation and communications — could become matters of national defense. But in peacetime these military needs diminished, and these aspects largely returned to civilian use.

Diplomacy had nothing to do with “national defense” but instead belonged to an entirely distinct concept: “foreign relations.” And secrecy had been mostly unimportant in American diplomacy in the decades before Wilson’s presidency.

An observer defined the term “national defense” in 1927: “National defense means the protection of the country, primarily, from outside aggression, international war, not internal disturbances for which we have police forces and constabulary.”

Of course, the Espionage Act was used against “internal disturbances.” In 1917-1918, it became a notorious government weapon. But the provision the government used wasn’t the one related to government secrets. Instead, the government abused a different section of the law, which prohibited “willfully obstruct[ing] the recruiting or enlistment service of the United States” during wartime, to punish people agitating for labor protections, doing journalism or refusing to engage in patriotic acts.

The main secrecy provisions of the Espionage Act, on the other hand, went largely unused at first. The earliest interwar prosecutions came with a trio of cases in the late 1930s, which involved the sharing of naval secrets with foreign countries. One case reached the Supreme Court in 1941 — the only time the court has addressed the secrecy provisions of the Espionage Act. The court defined the term “national defense” only briefly as “referring to the military and naval establishments and the related activities of national preparedness.” The court did not elaborate, because it did not need to: Americans still understood the concept just as they had in 1917. “National preparedness” merely referred to the military’s broader economic needs. But the Court’s definition is generally not understood today, because the concepts of “national defense” and “national preparedness” have both been forgotten.

Americans forgot these concepts because in the mid-1940s, a new idea, one theorized by a Princeton professor, took the country by storm: “national security.” That professor, Edward Mead Earle, loathed “national defense.” The older concept’s narrow military emphasis rendered it far too small, he thought. Thinking in terms of “defense” encouraged “sitting back and waiting until the enemy is at one’s gates,” he wrote.

This new idea of national security, as historian Dexter Fergie argued, “heralded a novel way of imagining the world.” National security was elastic, and in comparison with the older concept, it was enormous. Before this new idea emerged, there existed, as Fergie wrote, “no concept that linked together so many disparate policy domains, from information and infrastructure to terrorism and trade.”

As “national security” rose in prominence, the original meaning of “national defense” was lost. By the 1970s, legal scholars examining the Espionage Act were already looking at “national defense” with befuddlement.

In the second half of the 20th century, the secrecy provisions of the Espionage Act became a regular tool to prosecute those accused of providing government secrets to foreign countries — such as Julius and Ethel Rosenberg, who were executed in 1953 after being convicted of giving top-secret military information to the Soviets during World War II.

The use of the act to try to punish leakers, however, was extremely rare. The first attempt came in 1973, with charges against Daniel Ellsberg for leaking the Pentagon Papers. The case collapsed because of prosecutorial misconduct. Two other cases, both for leaking classified military information, subsequently succeeded in 1988 and 2003.

But by that point, people ceased to remember what “national defense” originally meant and instead increasingly treated it as synonymous with “national security.” This conflation reached a zenith in 2006, when a district judge wrongly defined “national defense information” as information that “could threaten the national security of the United States.”

The scope of the Espionage Act has accordingly exploded. A turning point came with the Obama administration, which prosecuted at least eight leakers under the Espionage Act. The Trump administration prosecuted at least another four. Many of these prosecutions have had nothing to do with military secrets. Julian Assange’s leak of U.S. diplomatic cables, which accounts for four of the 17 charges against him, certainly had nothing to do with the military. Reality Winner’s leak of an NSA document about the 2016 election had no clear military implications, nor did Terry Albury’s leak of mere law enforcement material. Joshua Schulte currently faces Espionage Act charges, even though the CIA cyber material he leaked has no evident military connection.

Understanding this older concept raises crucial questions of fundamental fairness. The concept of “the national defense” no longer meaningfully exists. How can we have prosecutions under a law with a dead concept at its core — a concept that defendants and juries cannot intuitively comprehend?

The Espionage Act has become so dangerous only by a forgetting of history. The act originally was a vigorous but narrow military secrecy law — not the leviathan of government secrecy it has become.