This tribute is part of a series of reflections from HLS faculty, staff and alumni on the 15th anniversary of 9/11. View all tributes.

Contributor: Philip Heymann ’63

Harvard Law School Professor Philip Heymann ’63 has served at high levels in both the State and Justice Departments during the Kennedy, Johnson, Carter, and Clinton administrations including Deputy U.S. Attorney General (1993-1994). He co-wrote “Preserving Liberty in an Age of Terror” (MIT Press 2005), with Juliette Kayyem ’95, and “Laws, Outlaws, and Terrorists” (MIT Press, 2010), with Harvard Law Professor Gabriella Blum LL.M. ’01 S.J.D. ’03.

The massive Al Qaeda attacks fifteen years ago were planned and carried out by people from outside the country. That gave rise to a set of extremely troublesome legal issues. Was this “war?” Whether or not it was, what should be the legal requirements dealing with harsh interrogation, detention without trial, military commissions, and surprising new forms of surveillance? These have now been largely resolved; the most difficult legal challenges fifteen years later arise from domestic sources.

The perpetrators of the terrorist attacks in Paris, Brussels, Orlando, and Boston were all on some form of terrorist “watch list.” Although regarded as a danger, the government could not, it generally explained, afford to surveil the suspect’s activities over a long period. He was one of many and each would require many officers for full time surveillance.

This dilemma sets the stage for the hardest issue in addressing domestic terrorism. Assume that, on the basis of informant information about his hateful rantings or of other aspects of his behavior, an individual suspect has been the subject of a limited preliminary investigation by the FBI or one of its foreign or domestic counterparts and that his family, friends, and workmates have been interviewed.

Assume, as well, that the explicit or implicit conclusion of that investigation is that there is a 5% or 10% or 45% chance that he will, before long, attempt a lethal terrorist attack within the United States. Major Nidal Hassan, perpetrator of the killings at Fort Hood, would have been somewhere in this category. So would the older Tsarnaev in Boston after the U.S. received information from Russia. What can law enforcement or intelligence agents do to prevent that future attack?

“A critical part of dealing with domestic terrorism is to reduce the likelihood of an attack even if it cannot be eliminated. If there are non-punitive steps that could be taken legally to reduce, however much, the chance of the suspect engaging in terrorism, we should take those steps.”

The suspect cannot be arrested for a crime unless there is probable cause that he satisfies the required elements of an attempt or conspiracy or material support, including a step confirming the solidity and clarity of his intent to engage in terrorism. Even a 45% estimate of danger based only on his expressions of rage would not justify his arrest. And detaining a suspect without criminal charges is wrong. Our deep democratic suspicion of detention of mentally sound citizens without criminal charges is a very fundamental pillar of our democracy. That commitment was heightened for U.S. citizens by shame at the internment of Japanese/Americans during World War II.

We could try to make up the deficit of proof for an arrest in either of two ways. We now use informants and undercover officers to encourage the forward movement of a terrorist plan in order to reach the level of evidence of commitment required for charging or arresting for an attempt or conspiracy or material support. That option can only be useful if the suspect is taken in by the deception; but, even then, it is too costly for its use to be much expanded. We might well be setting up a severe punishment for a crime that never would have been attempted; and if so the financial costs to the government of a prolonged prison sentence are immense and pointless.

Alternatively, we could monitor the suspect for evidence that he has taken legal steps to acquire lethal weapons (such as automatic guns, explosives, and poisons). But if the weapons were dual-purpose—like Tsarnaev’s firecrackers and pressure cookers in Boston, Timothy McVeigh’s fertilizer in Oklahoma City, or Omar Mateen’s weapons acquired as a security guard in Orlando—that acquisition might still not have made an attack likely enough for an arrest for attempt, conspiracy, or material support.

Forgetting about arrest or detention of the suspect, officers could be authorized to bring before a court evidence of the grounds for believing that the suspect is likely to engage in terrorism, arguing that he therefore should be denied the right to purchase especially dangerous equipment. The proceeding might be triggered by a requirement for sellers of such equipment to check a watch list. But those not on a watch list could be used as agents for the suspect’s purchase. Alternatively, officers could try to either protect or simply keep the suspect away from any likely terrorism target. Yet there are far too many potential “soft” targets to identify “likely” ones.

Evidence of acquisition of potentially lethal equipment or of “casing” a target site or of seeking access to it or of recruiting co-conspirators significantly increases the overall likelihood of the suspect’s engaging in terrorism (beyond whatever probability arises from evidence of angry or threatening words). That greater likelihood could be used to meet a legislatively reduced standard for electronic or other surveillance and to search private data sources to identify associates who could be interviewed. Alternatively, the additional likelihood could be used systematically to re-interview the suspect’s family, close friends, or confidants.

The best alternative to these options is not simply to drop the matter if an early initial investigation does not produce enough evidence for an arrest. A critical part of dealing with domestic terrorism is to reduce the likelihood of an attack even if it cannot be eliminated. If there are non-punitive steps that could be taken legally to reduce, however much, the chance of the suspect engaging in terrorism, we should take those steps. Examples include urging a suspect’s family to discourage his plan in order to avoid his serving a very long sentence; or making clear to the possible terrorist that he will be a prime suspect if terrorism occurs; or attempting to ameliorate with him the causes of his anger against his country; or engaging in regular periodic conversations with his close friends; and more. This might not be a job for the FBI. A joint state-federal task force or Homeland Security may be preferable. But, if it is one of the latter two, the handoff from the FBI to a successor must be seamless and information must continue to flow freely in both directions.