FAN 156 (First Amendment News) Special Post: The Espionage Act at the 100 Year Mark: Commentaries by Bambauer, Chemerinsky, Stone & Vladeck
There are citizens of the United States . . . born under other flags but welcomed under our generous naturalization laws to the full freedom and opportunity of America, who have poured the poison of disloyalty into the very arteries of our national life; who have sought to bring the authority and good name of our Government into contempt, to destroy our industries wherever they thought it effective for their vindictive purposes to strike at them, and to debase our politics to the uses of foreign intrigue. . . .
I urge you to enact . . . laws at the earliest possible moment and feel that in doing so I am urging you to do nothing less than save the honor and self-respect of the nation. Such creatures of passion, disloyalty, and anarchy must be crushed out. — Woodrow Wilson, State of the Union Address, December 7, 1915
[T]he newspaper or individual who criticizes or points out defects in policies . . . with the honest purpose of promoting remedial action and warning against danger is not a public enemy. — Editorial, New York Times, April 13, 1917
Today marks the 100th anniversary of the Espionage Act of 1917 (18 U.S. Code Chapter 37). In light of that, I have collected some background materials about the Act followed by several original comentaries on it, which follow the introductory materials below.
- Assistant Attorney General Charles Warren drafts a bill “for suppressing or punishing disloyal and hostile acts and utterances.”
- Bill introduced in the House as H.R. 291
- Bill passes in the House on May 4, 1917 (261–109)
- Bill passes the Senate on May 14, 1917 (80–8)
- President Woodrow Wilson signs bill into law on June 15, 1917.
→ First Amendment Online Library Timeline of Espionage Act & Related Acts
→ The Masses cases:
- Masses Publishing Co. v. Patten, 244 F. 535 (S.Dist.N.Y., 1917) (per Hand, J.)
- ruling re stay of appealed order, 245 F. 102 (per Hough, J., 1917), and
- Circuit Court ruling reversing District Court, 246 F. 24 (2nd Cir., 1917) (per Rogers, J. for the majority with Ward, J. concurring)
Controversial Provisions of the Act: “The Espionage Act put into law a penalty of up to 20 years imprisonment for anyone convicted of interfering with military recruitment. The law also presented the penalty of levying fines of up to $10,000 for those convicted. The law also gave additional powers to the post office. Specifically, the law allowed the Postmaster General to confiscate any mail that might be deemed seditious or treasonable.” (source: This Day in History)
Domestic Issues of Concern: “There were quite a number of concerns the Wilson administration had about certain groups that were in opposition to the war. Public criticism of the war was definitely a major concern of the government. Since a significant number of troops would be needed to carry out the war effort, a draft was imposed. Among the concerns the government had was the notion that constant criticism would make recruitment and even conscription difficult.” (source: This Day in History)
Enforcement: “Enforced largely by A. Mitchell Palmer, the United States attorney general under President Woodrow Wilson, the Espionage Act essentially made it a crime for any person to convey information intended to interfere with the U.S. armed forces prosecution of the war effort or to promote the success of the country’s enemies. Anyone found guilty of such acts would be subject to a fine of $10,000 and a prison sentence of 20 years.” (Source: Totally History)
→ See also David Greene, As the Espionage Act Turns 100, We Condemn Threats Against Wikileaks, Electronic Frontier Foundation, June 14, 2017
→ Zechariah Chafee, Jr., Freedom of Speech in War Time, 32 Harvard Law Review 932 (1919)
→ Karl N. Llewellyn, Free Speech in Time of Peace, 29 Yale Law Journal 337 (1920) (student comment)
→ Walter Nelles, In the Wake of the Espionage Act, The Nation (December 15, 1920)
- Geoffrey Stone, Learned Hand and the Espionage Act of 1917: A Mystery Unraveled, 70 University of Chicago Law Review 335 (2003)
- Geoffrey Stone, Perilous Times: Free Speech in Wartime (W.W. Norton, 2004), pp. 146-153, passim
- Stephen Feldman, Free Expression & Democracy in America: A History (University of Chicago Press, 2008), pp. 241-245 passim
Commentaries by Derik Bambauer, Erwin Chemerinsky, Geoffrey Stone & Stepehen Vladeck
Backwards and Forwards
Many thanks to Ron Collins for the invitation to reflect on the centennial of the Espionage Act!
I want to argue that the Espionage Act is not only problematic on its own terms, but that it has paved the way for a newer set of worrisome statutes and dubious cases. These newer measures, like the Act itself, respond to an exaggerated sense of danger from internal and external threats. Courts and lawmakers alike have largely engaged in ahistorical analysis: they have failed to learn the lessons from the past. They underrated the perceived risk of subversive political speech in wartime in the early twentieth century, and they fail to question whether terrorism is a sufficiently existential threat today to warrant impingements on speech. I conclude that there are two possible responses, neither particularly promising.
To give the problems with the Espionage Act and its progency some currency, consider the current fight against the terror group ISIS in the Middle East. ISIS has taken a surprising amount of territory, and has become infamous for its horrific treatment of captives. There have been domestic attacks by individuals or small groups who claimed an association with or allegiance to ISIS. However, while these attacks are horrifying, even significant terror attacks are not existential threats to the United States.
But the perceived threat from ISIS has generated a disproportionate response, and one that frequently targets speech. For example, the assistant attorney general for national security said that people who are “proliferating ISIS social media” could be prosecuted under 18 U.S.C. 2339A (the material support statute). His contention was that disseminating ISIS’s point of view counts as providing “technical expertise” to a terrorist group. In 2012, the government successfully prosecuted Tarek Mehanna as a terrorist, in part because he translated al Qaeda writings and videos into English. The First Circuit Court of Appeals, in affirming his conviction, called terrorism the “modern-day equivalent of the bubonic plague” and an “existential threat” – an embarrassing example of hyberbole. And the Supreme Court has failed to rein in restrictions on speech justified as necessary for the war on terror. In Holder v. Humanitarian Law Project, the Supreme Court rejected a First Amendment challenge to the material support statute, highlighting the fact that completely independent political advocacy is not covered by the law’s prohibitions. (The challenge, of course, is determining when someone is “completely independent.”)
The material support statute is also problematic in that it defers decisions about what content should be criminalized to the executive branch. The State Department is empowered to determine which entities constitute terrorist groups. Coordinated political advocacy with groups on the list is a crime; advocacy for violent but not listed groups is safe. One person’s terrorist is another’s freedom fighter. There’s also the risk of one-way advocacy: if terrorist group calls for people to advocate on its behalf, and someone does so, does that count as coordination? And, of course, the Justice Department has charged Edward Snowden under the Espionage Act itself, and continues to investigate whether to prosecute WikiLeaks and Julian Assange under it.
These efforts seem similar to prosecutions in the early twentieth century under the Espionage Act that ultimately elucidated the weak form of the “clear and present danger” test. For example, Charles Schenck was convicted for distributing pamphlets for American Socialist Party that read “Assert your rights – do not submit to intimidation” – hardly stuff to stir the blood. The Socialist Eugene Debs was convicted for giving a speech titled “Socialism is the Answer.” And Jacob Abrams was an anarchist convicted for his criticism of the U.S. decision to defend Russia against the Bolsheviks.
There were similar trends during the Cold War. The Communist Eugene Dennis received his conviction for knowingly advocating the overthrow of the U.S. government by force. The Supreme Court called the Communist Party a “permanently organized, well-funded, semi-secret organization.” These cases demonstrate at least two parallels to current events. The first is a conception of political (largely foreign) enemies as an existential threat. The second is that the magnitude of this threat justifies restrictions on political speech and advocacy that would ordinarily be at the heart of First Amendment protection.
I can see two tentative responses to these problems, one pessimistic and one mildly optimistic.
The pessimistic channels Geoffrey Stone: we must accept as inevitable that there will be limits on political advocacy, especially in wartime, even if that war is a long twilight struggle rather than a declared conflict. This has the interesting side effect of making First Amendment less exceptional – there turns out to be an implicit balancing test even with “core” political speech. It might also be a useful descriptive exercise to examine the pendulum swing of First Amendment liberties – is it temporal in nature, or does it relate primarily to subject matter?
The optimistic idea is to draw upon the historical parallels elaborated above: neither socialists nor Communists proved an existential threat to U.S. politics and institutions, despite the heated fears of the moment. This requires more work on our collective part: we have to hold the government to its burden when it seeks to restrict speech, first by questioning the characterization of information as a threat, and second by carefully policing the line between conduct and speech when regulations are proposed or promulgated. That will require political courage – always in short supply – and legal analysis grounded in history.
The Espionage Act ought to teach us that these are vital assets in our self-governance, but it’s not clear we have yet learned the lesson.
A Loaded Gun
The Espionage Act of 1917 is a loaded gun waiting for the federal government to use it to punish speech. Indeed, throughout its history, it has been used to punish speech that should be deemed constitutionally protected. I especially worry that the Obama administration has set a precedent for the Trump administration, which has shown great hostility to the press. Since the enactment of the Espionage Act of 1917, twelve prosecutions have been brought under it for disclosures of information and nine of those were during the Obama administration.
The Espionage Act, by its very terms, is directed at restricting speech. The law makes it a crime to convey information with the intent to interfere with the operation of the armed forces or to promote the success of its enemies. The Act also makes it a federal crime to convey false reports or false statements with intent to interfere with the operation or success of the military or naval forces of the United States or to promote the success of its enemies when the United States is at war, to cause or attempt to cause insubordination, disloyalty, mutiny, refusal of duty, in the military or naval forces of the United States, or to willfully obstruct the recruiting or enlistment service of the United States. Yet another provision gives the Postmaster General the authority to impound or to refuse to mail publications that he determined to be in violation of its prohibitions.
The Act has been used to punish speech. Most famously, it was used to punish speech during World War I that the First Amendment never should have allowed to be punished. In Schenck v. United States, the Court considered the conviction of two individuals – Charles Schenck and Elizabeth Baer – who were prosecuted for circulating a leaflet arguing that the draft violated the Thirteenth Amendment as a form of involuntary servitude. The leaflet was titled, “Long Live the Constitution of the United States.” It said, ‘‘Do not submit to intimidation,’’ and ‘‘Assert Your Rights,’’ but did not expressly urge violation of any law; it advocated repealing the draft law and encouraged people to write to their representatives in Congress to do so.
There was not any evidence that their leaflet had any effect in causing a single person to resist the draft. Nonetheless, they were prosecuted and convicted and sentenced to jail for violating the 1917 Act. The Supreme Court, in an opinion by Justice Oliver Wendell Holmes, upheld their convictions and sentences; he dismissed as irrelevant that the leaflet had no effect.
A week after Schenck was announced, the Court upheld convictions under the 1917 Act in two other cases, Frohwerk v. United States and Debs v. United States. Jacob Frohwerk was the publisher of a German language newspaper, Missouri Staats-Zeitung. He was prosecuted for a dozen articles published between June and December 1917. Again, the speech was the antithesis of shouting fire in a crowded theater or that which would pose a clear and present danger. Without doubt, any court today would regard it as expression protected by the very core of the First Amendment. It took the jury only three minutes of deliberation to convict Frohwerk of violating the 1917 Espionage Act and the judge sentenced him to 10 years in prison for his writings. The Supreme Court affirmed.
In Debs v. United States, the Court affirmed the conviction of Socialist Party leader Eugene Debs who had been sentenced to jail for ten years for violating the 1917 Act. Debs was a national political figure, having run for President in 1900, 1904, 1908, and 1912. Debs was convicted for a speech that was primarily advocacy of socialism, but it included some mild criticism of the draft. At one point in a long speech, Debs remarked that he had to be ‘‘prudent’’ and not say all that he thought, but that ‘‘you need to know that you are fit for something better than slavery and cannon fodder.’’ For this mild statement Debs was convicted of attempting to incite disloyalty in the military and obstruct the draft. Again, the Supreme Court affirmed.
Moreover, during World War I, the Postmaster used the authority under the statute to seize magazines and newspapers. Albert Burleson, a reactionary racist from Texas who despised labor unions and the people who supported them, began a campaign to root out magazines and newspapers that promoted socialist or radical causes.
The subsequent use of the Espionage Act reinforces reasons for great concern. It is the statute used to prosecute Daniel Ellsberg and Chelsea Manning. It is a law that can be used to punish those who provide information to the press and to those who disseminate information. The Act is so broadly written that there is no way to know the speech that it has chilled over the years. I have great fears of how it might be used in the next four years with a President who has shown such great hostility to the press.
When Can a Government Employee Leak Classified Information?
The Espionage Act of 1917, as amended over the years, forbids government employees to disclose classified information to any person who is not authorized to have access to it. For this reason, most prosecutions of government leakers of classified information have relied upon the Espionage Act. The Act recognizes no defense for government employees who leak such information. This is one of Edward Snowden’s justifications for refusing to return to the United States to face prosecution. He maintains that the absence of a defense that would exonerate government leakers of classified information whose acts do more good than harm is unjust. As we look to the future, a central question is whether the Espionage Act should be amended to recognize such a defense.
At first blush, there is obvious logic in Snowden’s position. After all, if someone does more good than harm, shouldn’t they be free to do the good? Moreover, this seems especially sensible in the context of classified information, because the test for classification is whether the disclosure of the information might “reasonably be expected to harm the national security.” There is no balancing at all of good versus harm. The standard does not take into account the possible benefits of the leak and it does not require that the harm be likely, imminent, or grave.
Moreover, the government quite predictably tends to over-classify information. The simple rule is: Better be safe than sorry. In addition, we know from experience that public officials have on occasion abused the classification system in order to hide from public scrutiny their own misjudgments, incompetence, and venality.
In light of these concerns, it might seem logical to amend the Espionage Act to permit a government employee legally to disclose classified information whenever she can demonstrate that the benefit of the disclosure outweighed the actual harm to the national security. Why shouldn’t Edward Snowden have such a defense, if he can prove the case? After all, granting such a high level of deference to the government in these situations significantly overprotects government secrecy at the expense of both official accountability and informed public debate. Even worse, in some situations the leaker might disclose the existence of programs that are themselves unlawful. In that case, how can it possibly be right to make it a crime for the government employee to disclose the information to the public?
The government’s response to all this is fairly straightforward. First, except in extraordinary circumstances like self-defense, we don’t give individuals a right to break the law because, in the circumstances, committing the crime might do more good than harm. For example, if X steals someone’s purse because he needs money to feed his children, he could easily argue that his theft did more good than harm, but that is not a defense. One could, of course, multiply that hypothetical endlessly.
Second, there are more than a million government employees and private contractors who have access to classified information. The government will argue that it would be reckless in the extreme to permit each of those individuals to think that it is permissible for them to disclose classified information whenever they conclude that the good would outweigh the harm. Even if in some instances they might be right, often they will be wrong – especially because individual government employees and contractors are rarely in a position to understand how the information they plan to disclose might damage the national security. Thus, the government will argue, the only sensible thing to do is to take that option away from these employees. Finally, the government will point out that in order to prove in court that a leak caused substantial damage it would often have to reveal even more classified information, often including sources and methods, which would make such inquiries especially problematic.
So, what’s to be done? It seems unlikely that the Supreme Court will recognize a First Amendment right of government employees to leak classified information. The implementation of a constitutional rule that permits leaks would just be too messy for the Court to impose or to implement. As a legislative matter, though, it would make sense to create some internal mechanism through which these employees can raise their concerns, especially if they believe the programs at issue to be unlawful. To-date, though, there seems to be little interest in such an option. Another possibility, of course, is simply to tighten up the standards and procedures for classification. No one doubts that we currently live in a world of gross over-classification.
Are there cases one can imagine in which even under existing law it would seem implausible to punish a leaker? Suppose an FBI agent learns from a classified document that at the direction of the Russian government the FBI assassinated the president? I rather suspect that if she leaked that information, assuming it is accurate, she would not go to jail.
It’s (Long-Past) Time to Modernize the Espionage Act
For a law that turns 100 today, and that’s only been materially amended once in a century, the Espionage Act has sure enjoyed a popular resurgence of late. President Barack Obama used it to prosecute more leakers than all previous Presidents put together. Critics of Secretary Hillary Clinton’s unauthorized use of a private e-mail server sought desperately to make the (legally unconvincing but politically damaging) argument that she had violated the statute. And when former FBI Director Jim Comey revealed just last week that he had been responsible for leaking a memo memorializing a conversation he had with President Trump, commentators quickly gravitated toward the Act as proof that, if any laws were broken as part of Comey’s termination, it was the old chestnut herself through Comey’s transgression.
What all of these recent stories have in common is the absence of actual “espionage”— the conduct that motivated Congress to enact the law in the first place. The Act was written on the eve of the United States’s entry into World War I, and, as importantly, before the emergence of either the modern terminology for national security classification or the Supreme Court’s modern First Amendment and vagueness jurisprudence. But because Congress has resisted decades of calls to revisit it, it remains on the books mostly as initially enacted—a statute aimed at German spies. Its clunky and capacious language paints with the same brush three distinct offenses:
- classic espionage,
- leaking, and
- the retention or redistribution of national defense information by third parties.
Part of why it’s so problematic that the Espionage Act treats these three very different sins as the same crime is because of its outdated (and outmoded) language. We now have a sophisticated series of Executive Orders that define and regulate the scope of “classified” national security information, and that have, unlike the Espionage Act, regularly been updated to respond to changes in technologies and threat vectors. One would think it is those provisos, and not a century-old statute, that better reflect the true contemporary scope of “information relating to the national defense.”
And, along with classification, we also now have a far-more-sophisticated understanding of the problem of over-classification—and why it’s so problematic that courts have refused to recognize “improper classification” as a defense to an Espionage Act prosecution. Thus, the more information that has become classified, the easier it has become to violate the Espionage Act through conduct that bears increasingly less resemblance to spying.
Finally, we also now have First Amendment jurisprudence that recognizes at least some circumstances (albeit virtually none thus far involving national security information) in which the disclosure of certain previously confidential material might be of such surpassing public concern as to be protected by the First Amendment even when it might otherwise be unlawful. But the 100-year-old verbiage of the Espionage Act doesn’t account for any of these developments. That may be why, 37 years ago (before it qualified for Medicare), the Act was decried by Anthony Lapham, then the General Counsel of the CIA, as the “the worst of both worlds.” As he then explained to Congress:
On the one hand the laws stand idle and are not enforced at least in part because their meaning is so obscure, and on the other hand it is likely that the very obscurity of these laws serves to deter perfectly legitimate expression and debate by persons who must be as unsure of their liabilities as I am unsure of their obligations.
Modernizing the Espionage Act won’t be easy. But 100 years in, it’s long past time for Congress to do so.