Can the government compensate for its failure to keep its secrets secret by punishing those who pass the leakage on to the public in a manner judged "prejudicial" to national interests?  Not, says constitutional law scholar Geoffrey R. Stone in a New York Times guest column, under the First Amendment.

The so-called Shield bill, which was recently introduced in both houses of Congress in response to the WikiLeaks disclosures, would amend the Espionage Act of 1917 to make it a crime for any person knowingly and willfully to disseminate, “in any manner prejudicial to the safety or interest of the United States,” any classified information “concerning the human intelligence activities of the United States.”

Although this proposed law may be constitutional as applied to government employees who unlawfully leak such material to people who are unauthorized to receive it, it would plainly violate the First Amendment to punish anyone who might publish or otherwise circulate the information after it has been leaked. At the very least, the act must be expressly limited to situations in which the spread of the classified information poses a clear and imminent danger of grave harm to the nation.

The clear and present danger standard has been a central element of our First Amendment jurisprudence ever since Justice Oliver Wendell Holmes Jr.’s 1919 opinion in Schenk v. United States. In the 90 years since, the precise meaning of “clear and present danger” has evolved, but the animating principle was stated brilliantly by Justice Louis D. Brandeis in his 1927 concurring opinion in Whitney v. California. The founders “did not exalt order at the cost of liberty,” wrote Brandeis; on the contrary, they understood that “only an emergency can justify repression. Such must be the rule if authority is to be reconciled with freedom. Such … is the command of the Constitution. It is, therefore, always open to Americans to challenge a law abridging free speech and assembly by showing that there was no emergency justifying it.”

On the other hand, the First Amendment does not compel government transparency. It leaves the government extraordinary autonomy to protect its own secrets. It does not accord anyone the right to have the government disclose information about its actions or policies, and it cedes to the government considerable authority to restrict the speech of its own employees. What it does not do, however, is allow the government to suppress the free speech of others when it has failed to keep its own secrets.

We might think of this like the attorney-client privilege. If a lawyer reveals his client’s confidences to a reporter, he can be punished for violating that privilege — but the newspaper cannot constitutionally be punished for publishing the information.

There are very good reasons why it makes sense to give the government so little authority to punish the circulation of unlawfully leaked information.

First, the mere fact that such information might “prejudice the interests of the United States” does not mean that that harm outweighs the benefit of publication; in many circumstances, it may be extremely valuable to public understanding. Consider, for example, classified information about the absence of weapons of mass destruction in Iraq.

Second, the reasons that government officials want secrecy are many and varied. They range from the truly compelling to the patently illegitimate. As we have learned from our own history, it is often very tempting for government officials to overstate their need for secrecy, especially in times of national anxiety. A strict clear and present danger standard — rather than an unwieldy and unpredictable case-by-case balancing of harm against benefit — establishes a high bar to protect us against this danger.

And finally, a central principle of the First Amendment is that the suppression of free speech must be the government’s last rather than its first resort in addressing a problem. The most obvious way for the government to prevent the danger posed by the circulation of classified material is by ensuring that information that must be kept secret is not leaked in the first place.

Indeed, the Supreme Court made this point quite clearly in its 2001 decision in Bartnicki v. Vopper, which held that when an individual receives information “from a source who obtained it unlawfully,” that individual may not be punished for publicly disseminating the information “absent a need … of the highest order.”

The court explained that if the sanctions now attached to the underlying criminal act “do not provide sufficient deterrence,” then perhaps they should be “made more severe” — but “it would be quite remarkable to hold” that an individual can constitutionally be punished merely for publishing information because the government failed to “deter conduct by a non-law-abiding third party.” This is a sound solution.

If we grant the government too much power to punish those who disseminate information, then we risk too great a sacrifice of public deliberation; if we grant the government too little power to control confidentiality at the source, then we risk too great a sacrifice of secrecy. The answer is thus to reconcile the irreconcilable values of secrecy and accountability by guaranteeing both a strong authority of the government to prohibit leaks and an expansive right of others to disseminate information to the public.