Criminalizing statements of terrorist intent: how to understand the law governing terrorist threats, and why it should be used instead of long-term preventive detention.
What ought we, as a liberal society, do with members of our society who have stated their intentions to commit terrorist attacks? Preventively detain them to ensure that they do not have the opportunity to act on those intentions? Monitor them with the goal of catching them, hopefully before they do any harm, in a criminal act for which they can be prosecuted and, if convicted, imprisoned? Or prosecute them for having stated that intention? I argue here for the last option, which can be pursued by prosecuting them for threatening to commit terrorist acts. (1)
The kind of threat with which I am concerned is not the paradigmatic kind of threat, in which the person making the threat tries to communicate the threat to one or more victims. (2) It is the kind of threat in which the potential victims may never know of the threat, but others, who may feel personally safe, have reason to believe that the person who made the threat intends to perform the threatened action. In other words, they think he is a threat by virtue of his making a threat.
This can easily be misconstrued as allowing the state to punish on the basis of a prediction of dangerousness. It is not that. Threatening, in the sense of stating one's intention to commit a violent criminal act, is an inchoate crime, (3) similar to, but more inchoate than attempt.
One might worry that prosecutions for threat would run up against the First Amendment protection of free speech. But the doctrine of "true threats" holds that while political hyperbole is protected speech, true threats are not. (4) The problem in this area of constitutional law is only that the case law discussing true threats muddles a number of important distinctions. I will argue, however, that the doctrine can be cleaned up to show why it does and should support prosecuting those who state their intentions to commit terrorist (or other violent criminal) acts.
I proceed in four parts. First, in Part II, I present a hypothetical but realistic case that illustrates the potential for this problem to arise. This example helps illustrate the basic concerns that point in favor of the prosecutorial option. In Part III, I summarize an argument I have made elsewhere, that U.S. citizens should not be subjected to long-term preventive detention, at least not now, when the police and the courts are functioning normally. (5) In Part IV, I explain briefly why monitoring alone is not a position that should be relied on if detention is a legitimate option. And in Part V, I discuss the crime of making threats, including terrorist threats. To address the incoherence in threat doctrine as it now stands, this Part is divided into three sections. In the first, I describe the doctrine of true threats and explain why it is incoherent. In the second, I explain how it should be carved up into two distinct crimes: threatening a victim (a potentially complete crime) and stating the intention to commit a crime (an essentially inchoate crime). In the third, I defend the inchoate crime against the objection that it takes the notion of inchoate crimes too far from completed criminal acts.
II. ILLUSTRATING THE ADVANTAGE OF USING THREAT CRIMES
Consider this hypothetical. The CIA receives a tip from U, an undercover agent in Pakistan, that an American citizen, A, is working with members of al Qaeda on a plot to set off a bomb in Los Angeles. The FBI uses that information to get a wiretap on A under the Foreign Intelligence Surveillance Act (FISA). (6) They listen for a few months and note that she has made a number of calls to people who are suspected of being al Qaeda members in Pakistan and the U.S., but the content of the calls is not obviously incriminating. U is consulted, and he says that discussions about charity work in the U.S. are really coded discussions of planned attacks. In that light, the discussions are highly incriminating. If the Government could make the case that A has been plotting to attack Los Angeles, it could prosecute A for conspiracy to commit terrorism. (7) Without calling U in to testify to explain how he knows what the code is and to provide other substantiating information, however, the recorded phone calls would not support proof beyond a reasonable doubt that A is conspiring to commit a terrorist act. Moreover, U cannot be brought in to testify against A without blowing his cover, something the U.S. is currently unwilling to do. (8)
The U.S. Attorney for Los Angeles has looked into prosecuting A for providing material support to a foreign terrorist organization (FTO). (9) But there is no evidence that she has provided money or services to an FTO. As a next move, the FBI has tried to set up a sting operation, offering to help A obtain explosives, but A consistently, even indignantly, declines. This could be because she is not the terrorist that U says she is, or it could be that she is sufficiently networked into a real terrorist organization that she does not need, and is in fact very careful not to work with people who have not been carefully vetted by her or other al Qaeda members she trusts.
What should the government do? The traditional answer from civil libertarians is that the government must limit itself to continuing to police A, trying to apprehend and prosecute her if and when she takes action that would constitute a crime, hopefully before she causes any harm. (10) That is, the government has to hope it can catch her attempting or conspiring to commit a terrorist act, providing material support to terrorist organizations, or committing some other crime that does not involve completing a violent terrorist act. (11) On the other side of the spectrum, those who take the idea that the U.S. is at war with al Qaeda most seriously would call for preventively detaining A as an enemy combatant (EC). (12) Even President Obama, for all of his respect for the rule of law, seems pulled to this side of the spectrum. After saying that "[w]e are going to exhaust every avenue that we have to prosecute those at Guantanamo who pose a danger to our country," he admitted that "there may be a number of people who cannot be prosecuted for past crimes, but who nonetheless pose a threat to the security of the United States." (13) He included in that list those who "made it clear that they want to kill Americans." As he said, "[t]hese are people who, in effect, remain at war with the United States." (14) Granted, he was talking about aliens detained in Guantanamo, not U.S. citizens. But his logic could easily be extended to cover U.S. citizens as well. No doubt, any responsible administration following that logic would want to ensure that U.S. citizens who seem to be members of or collaborators with al Qaeda would have a fair opportunity to contest their designation as ECs. (15) The point, however, is that these procedures would be much more flexible--with more room for hearsay and a lower burden of proof--than the procedures used in either civilian trials or military commissions. (16)
I have argued elsewhere that the government could be justified in putting U.S. citizens in short-term preventive detention, (17) but not long-term preventive detention. Short-term detention is not such a great burden that we cannot expect the innocent to endure it, while the government tries to determine whether they are in fact innocent, for the sake of national security. For example, we subject people who are awaiting trial, and who therefore have not yet been proven guilty beyond a reasonable doubt, to short-term preventive detention if they are considered dangerous. (18) Moreover, short-term detention has some obvious benefits, both in allowing the government to interrogate terrorism suspects, and in disrupting suspects' plans and relationships with other terrorists. (19)
Long-term detention--the line is hard to draw, but in many different legal contexts six months seems to mark a transition from short-term to long-term detention (20)--by contrast, is a great burden on an individual. It is more, I argue in Part III, than we can ask an innocent person to endure for the safety of others. Even if the detention is not meant to be punitive, the effect is sufficiently severe that an autonomous U.S. citizen (21)--one who has reached a threshold capacity to use practical reason to frame and pursue a conception of a good life (22)--may be subjected to it only if he has been convicted, beyond a reasonable doubt, of a crime for which a long-term sentence or loss of the right not to be subjected to long-term preventive detention is a fitting punishment. (23) If there is reasonable doubt that he is guilty--and it is hard to eliminate such doubt unless and until the defendant has had a fair opportunity to confront the evidence against him, something he may not be able to do if certain witnesses are not available--then I stand with the civil libertarians and say that such people must not be subjected to long-term preventive detention.
Let us now add the dimension of a threat. Suppose that a fellow member of A's mosque, M, comes forward and tells the FBI that A has told him that she, A, is a jihadist who plans to detonate a bomb in Los Angeles. (24) Does this make a difference? One might say that it is merely corroboration of what U has told the U.S. government. If M does not provide new information sufficient to convict A of either conspiring or attempting to commit terrorist acts, then he cannot help the government prosecute A. Indeed, even if A had been taken in by the FBI and, under interrogation, said the same thing to the FBI (25)--that she is a jihadist and that she plans to blow up a major building in Los Angeles--one might think that the civil libertarian position would require the government to let her go and wait for her to actually attempt or conspire to carry out her plan. (26)
That, however, would arguably be a reductio ad absurdum of the civil libertarian position. Surely the government should be able to detain A to ensure that she does not commit a terrible crime. Indeed, given the limited effectiveness of surveillance without detention--explored in Part III--longterm detention is really the only way to ensure that A does not commit terrorist acts. But if A's statements which do not mention her conspiring with others--merely corroborated what the government already believes, then the government still may not be able to prosecute A. And if the civil libertarian position implies that the government has no other option for long-term detention--if long-term preventive detention were off the table--then the civil libertarian position would be too extreme.
What A has said about her intentions is not, however, merely corroborative of what the government already believes. By stating her intention to commit terrorist acts, A commits a new crime, for which she can be prosecuted, the crime of threatening to commit a terrorist act. (27) Prosecuting her on that basis has the advantage of putting the government to the test of proving, beyond a reasonable doubt, that she did indeed threaten to commit terrorist acts--as opposed to joking, protesting, blustering, or performing some other speech act protected by the First Amendment--and that she deserves to be incarcerated for a long period of time as a result. (28)
Prosecuting A for a terrorist threat fits the civil libertarian commitment to avoiding long-term preventive detention. Moreover, the balance it strikes between liberty and security seems appropriate, given basic civil libertarian commitments. As I said above, if there is a reasonable doubt that A meant what she said, then the government should not take away her most fundamental liberty, her freedom from incarceration, (29) for a long period of time. This is not to impugn the credibility of U's information. But unless A has an opportunity to challenge U, U's saying that A is working with al Qaeda should not constitute proof beyond a reasonable doubt. The government's unwillingness to call on U to testify means that there could be no proof beyond a reasonable doubt of A's being guilty of a terrorism-related crime unless and until she did something more. The difference A's statements to M make, then, is not just that they corroborate what U said; they constitute a crime in their own fight. They allow the government to bring A to trial, to present M's evidence, to allow A to contest it, and to allow a jury to decide if A is guilty beyond a reasonable doubt.
In sum, prosecuting suspected terrorists like A for threatening to commit terrorist acts gives the government another important tool to fight terrorism. This tool may come into play in those cases that seem to provide the strongest case for using long-term preventive detention on U.S. citizens: those in which the government has good reason to believe that someone is plotting to commit a terrorist act but cannot, for whatever reason, (30) prove that beyond a reasonable doubt. If such a person sincerely states his intention to engage in terrorist acts, then that action by itself not only corroborates the view that he is dangerous, it is also a crime in its own right that gives the government a basis for a prosecution.
III. LIMITS ON USING PREVENTIVE DETENTION
One might respond to the claim that the government can prosecute suspected terrorists (STs) who threaten to commit terrorist acts by saying that that option, helpful as it might be, is not as helpful as also having the option of subjecting STs to long-term preventive detention (LTPD). I have argued at length against that position, (31) and in favor of the view that respect for the dignity of autonomous individuals requires the government to release and police, after (at most) a short period of preventive detention, its own citizens who it cannot convict of a crime for which long-term punitive detention is a fitting punishment. I am not alone in arguing for that sort of conclusion, (32) but I believe I am nearly alone in arguing for it on principled, rather than pragmatic, grounds. (33) I will summarize that argument here.
A. LEGAL AND MORAL BACKGROUND
Before starting, it is important to be clear that no one, myself included, has argued that the U.S., acting in cooperation with the Afghan government, cannot justifiably hold Taliban or al Qaeda fighters who participate in the insurrection against that government in LTPD. The insurrections or civil wars in places like Afghanistan, Pakistan, and Iraq create war zones in which the capacity of the state to police the residents of the territory is highly compromised. They are places where the conditions are of the sort which allow the suspension of the writ of habeas corpus in the U.S. (34) In such conditions, STs can be subjected to LTPD. The argument that follows concerns U.S. citizens whose release would be into the U.S. where, at least as I write this, the police and the courts are fully functioning.
My position is, however, inconsistent with the position adopted by the majority of the Supreme Court Justices writing in Hamdi v. Rumsfeld. (35) The plurality opinion, as well as the opinions of Justices Souter (joined by Ginsburg) and Thomas, all hold that it is constitutionally permissible for the U.S. to detain its citizens as ECs. (36) The plurality opinion based its decision in large part on Exparte Quirin, (37) which held that U.S. citizens can be tried as ECs for violations of the law of war. The plurality inferred that if U.S. citizens can be tried as ECs, then they can be detained as ECs as well. (38) But in my view Justices Scalia and Stevens, dissenting in Hamdi, were correct: the U.S. may not detain its own citizens as ECs. (39) Perhaps it may try them as ECs, if military commissions have jurisdiction over anyone who is accused of committing war crimes. (40) But ECs may not be subjected to LTPD simply because they are ECs. They may be preventively detained for long periods of time, I argue below, only if they cannot be held accountable. (41)
One central premise of my position is that war is not fundamentally different from peace; people have rights that must be respected in both conditions. Cicero said: "[D]uring war, the laws are silent" (silent enim legis inter arma). (42) But as Justice Barak, President (Emeritus) of the High Court of Justice in Israel, wrote: this "saying [is] regrettable. [It] reflect[s] neither the existing law nor the desirable law.... It is when the cannons roar that we especially need the laws." (43)
It is now clear that humanitarian law regulates war as a law-governed activity, one in which the law is informed by humanitarian moral concerns. (44) My moral point, however, goes beyond the mere relevance of humanitarian concerns. Not only does the law apply in war as in peace, its moral foundations are the same in war and in peace; in both, the law must respect people's basic rights. In saying this, I follow the many just war theorists who treat war as an activity within the reach of our normal moral concepts, one that should be regulated by appeal to normal moral reasons. (45) And it is in that spirit that I argue that all detentions must be justifiable within a moral framework that respects the liberty rights of autonomous individuals.
To be clear, I am not suggesting a model in which autonomy is a "value" to be promoted. That allows it to be merely one more good, even if a very important one, to be put in the balance with security. Rather, the model I rely on treats autonomy as a capacity that gives those who have it a kind of moral status that we might call dignity. Dignity is not a value to be promoted either; it is a status that must be respected. (46)
B. THE AUTONOMY RESPECTING MODEL AND ITS APPLICATION TO POWS AND STS
The liberal, dignity-respecting framework I am working with supports what I call the autonomy respecting model (AR model) of detention. Its core principle is that individuals who can be adequately policed and held criminally liable for their illegal choices as normal autonomous agents, and who can choose whether their interactions with others will be impermissibly harmful or not, can be subjected to long-term detention only if they have been convicted of a crime for which long-term punitive detention or the loss of the right not to be subjected to LTPD is a fitting punishment, or both.
One way to break this down into its component parts is as follows. Those who can be detained fall into two basic categories: those subject to punitive detention and those subject to preventive detention. Punitive detention respects autonomy because it is based on a person's autonomous choice to commit a crime. Those subject to preventive detention can be detained short-term for the sake of security, because even innocent people can be expected to make small sacrifices for the sake of the greater welfare. (47) People may properly be subjected to LTPD, however, only if they fall into one of four categories: (1) they lack the normal autonomous capacities to govern their own choices; (48) (2) they have, by virtue of one or more criminal convictions, lost their right to be treated as autonomous and accountable; (49) (3) they have an independent duty to avoid contact with others, because such contact would be impermissibly harmful, and LTPD simply reinforces this duty; (50) or (4) they are incapable of being adequately policed and held accountable for their choices. Importantly, prisoners of war (POWs) and some STs--I explain immediately below which STs--fit into this last category, and thus, their detention can be accounted for in the AR model. If, however, a given ST does not fit into this last category or any of the other categories just mentioned--and many STs do not--then he must either be tried and convicted of a crime or be released and policed like any criminal defendant who is acquitted at trial.
As the concern of this Article is with the detention of STs, I focus here only on the fourth of these categories. The basic thought is that there are two ways of being unaccountable: intrinsically and extrinsically. Those with insufficient autonomous capacity are intrinsically incapable of being held accountable as any normal autonomous person for their actions. (51) Those who cannot adequately be policed are extrinsically incapable of being held accountable, in anything like the normal way, for their actions. The moral relevance of both is the same: one who cannot adequately be held accountable can be subjected to LTPD if doing so is necessary to ensure that he does not pose a risk to others that outweighs the loss of his own liberty. This sort of balancing is appropriate when a person is not accountable because accountability is a precondition of treating someone as an autonomous agent who should be released and policed, rather than be subjected to LTPD.
POWs typically fit into this category of those who may be subjected to LTPD because the typical POW is, and will remain until the war is over or he is released from military service, privileged to engage in combat with the detaining power. (52) If he escapes, he has the right to take up arms again. (53) That means that the detaining power not only cannot hold him criminally responsible for his past violent actions--at least as long as those acts do not violate the laws of war that require him, for example, not to target noncombatants (54)--it also may not hold him criminally responsible for any future acts of violence that conform to the law of war. The state is not required to allow itself to be attacked. Therefore, it can subject POWs to LPTD to prevent them from attacking. And it can do so without disrespecting them as autonomous people because their legal status makes them unaccountable.
Some STs can also be justifiably subjected to LTPD for a related reason. On the assumption that the U.S. has no obligation to release and police alien STs in its own territory, (55) the question is: Would they be adequately policed if released to their home country or to some other country willing to take them? The answer in some cases--for example, Yemen (56)--is no. It is not that they have the legal status of being beyond criminal prosecution for future acts of terror. The problem is (a) that their legal status does not require the U.S. to release them in its own territory and police them there and (b) that where they do have a legal claim to reside, the chance that they would not be held accountable for any future acts of terror is too large. As a result, they are effectively unaccountable and can be subjected to LTPD without disrespecting them as autonomous people. (57)
What about LTPD for U.S. citizens who are also STs? There the argument regarding the state's obligation to police is different. The state exists to serve its citizens, so it cannot say to a citizen, as it can to an alien, that it will not accept the responsibility of policing him. Of course, even if a state has an obligation to police its own citizens, there may still be questions about whether it can adequately do so at a reasonable cost, given the technologies and the political situation at a given period of time. If not, then again LTPD may be justifiable because the person cannot be held accountable. But there is little reason to think that adequate policing is not almost always possible in the U.S., at least in the current condition in which no one argues that habeas can be suspended. (58)
Indeed, in the period since 9/11 the policing power of the U.S. has been quite successful at providing domestic security. (59) Admittedly, there have been some cases in which the system did not work properly, and it was only because of the relative incompetence of the would-be terrorists that the U.S. did not suffer a terrible attack. (60) Thus, one could easily imagine that the U.S. might have suffered one or more serious terrorist attacks since 9/11. But the problem in these cases is not that the U.S. government is allowing people who it knows are dangerous to roam freely; the problem is that it is unable to figure out who is dangerous. Therefore, it seems implausible that allowing citizen STs to be held in LTPD would provide a significant increase in overall security.
This is not to deny that a situation could arise in which a U.S. citizen could justifiably be subjected to LTPD as an ST. If an ST had particular knowledge or skills that made him distinctively valuable to a terrorist group trying to obtain weapons of mass destruction, and there was no way (a) to prevent him from passing on that information or using those skills while he retained his liberty to move about in society or (b) to prosecute him for a past crime, including threat crimes, and if the government nonetheless had sufficient reason to suspect him of having an interest in working with a terrorist group, then LTPD could be justifiable. But this theoretical possibility is, I believe, as unlikely to occur in reality as the ticking time bomb example that is so often cited to justify the theoretical possibility that torture might be justified. (61) I am not suggesting that the odds of a terrorist attack with a weapon of mass destruction are exceedingly low. (62) I am saying that the U.S. government either will not know that a particular individual poses an especially great threat, or, if it does know that, then it will be able to prosecute or adequately monitor him to prevent him from committing terrorist acts. As I argued in Part II, there might be a gap if the government could not prosecute those it comes to believe have the intention of committing terrorist acts because of things they say. But as long as terrorist threats can be criminalized, there should be no need to subject U.S. citizens to LTPD as STs. (63)
C. OBJECTION AND REPLY
One might object that my AR model cannot be relied on because its account of the LTPD of POWs is implausible. That is, one might claim that it is an essential feature of war that POWs can be detained until the cessation of active hostilities. (64) And one might object that this feature of war has nothing to do with respecting the dignity of autonomous people; it simply reflects the unique security needs that arise in war.
It turns out that it is possible to test whether this objection is sound. Under the laws of war, provided for in the Third Geneva Convention, a POW may renounce his status as a privileged combatant--an act called giving his parole--by giving his word that if released he will not again take up arms against the detaining power. (65) By giving his parole a POW changes his status such that if he is captured again, having violated his word, he is subject to criminal penalties. (66) The possibility of giving his parole at least opens up the legal possibility of policing a former POW's future actions. And this provides a test for the AR model because it opens up the possibility that a detaining country would have an obligation to release POWs whom it paroles.
This test may initially seem to speak against the AR model. Clearly the U.S. would not release all POWs who wanted to give their parole; doing so would be too dangerous. (67) In fact, however, parole presents no problem for the AR model. There are two reasons, consistent with the AR model, why the option of giving one's parole is not and should not be made widely available to POWs. And in the few cases that could conceivably fall outside those reasons, POWs should be entitled to give their parole and be released.
The first reason parole is not widely available is that the POWs' home countries may not allow their soldiers to accept it. (68) If their home countries will not allow them to give their parole, then it would be unreasonable for the U.S. to treat them as having abandoned their legal status as privileged to use force. The U.S. itself does not allow its soldiers to give their parole. The U.S. military code of conduct says, "I will accept neither parole nor special favors from the enemy." (69) Insofar as the U.S. is holding POWs from a country with a similar policy, then parole would not be an option; those POWs would be legally barred from acquiring a duty not to fight again. Thus, they would retain their privileged combatant status and, for that reason, would not be accountable for their future use of force.
Second, even if the enemy would allow its soldiers to give their parole, it would be peculiar and highly unusual to find an enemy who could be relied upon to enforce its soldiers' duties to keep their word. I am not denying that countries who allow their soldiers to be released on parole assume this duty. The Geneva Conventions are clear that "the Power on which [POWs released on parole] depend is bound neither to require nor to accept from them any service incompatible with the parole or promise given." (70) I am suggesting, however, that during war it would not be reasonable to expect the enemy to be scrupulous about enforcing this duty. The commentary to the Geneva Conventions supports this skepticism, noting that the text of the Geneva Conventions does not "specify the attitude to be taken by that Power, from the penal or disciplinary point of view, in regard to breach of parole by a member of its armed forces." (71)
Assuming that the U.S. cannot rely on its enemies to enforce a parolee's duty to stay out of the fight, this suffices to explain why POWs cannot simply give their word and regain their status as accountable agents. The U.S. can threaten to punish POWs whom it releases on parole and subsequently recaptures again using force against the U.S., but given that these former POWs would be outside the jurisdiction of the U.S., that would be a very tenuous form of accountability. In reality, POWs would effectively be released only on their own honor. (72) If they lack honor, there is little that the former detaining power can do about it. They can be held accountable in one sense: it would be just to punish them if they are caught, tried and convicted. But they are effectively unaccountable in another sense: they are very unlikely ever to be caught, tried and convicted. This makes their situation similar to STs from Yemen: (73) the effectiveness of policing to provide for security does not reach the threshold level required for the U.S. to be obliged to release them. That is why LTPD, in this context, would be justified. (74)
If, however, the U.S. could trust that an enemy would enforce the duty of parolees to stay out of the fight, then it is quite plausible to insist that it would be obligated to give POWs the option of accepting parole. For the U.S. cannot simply choose to treat someone as unaccountable--except insofar as it can choose not to assume responsibility to police aliens. LTPD can be justified only if the U.S. confronts the fact that a person is not accountable if released into any country obliged or willing to accept responsibility for him. Thus, under the supposition, unrealistic as it may be, that an enemy would enforce the duty of parolees to stay out of the fight, they would be held accountable for any failure to respect their duty, and the U.S. would then be obliged to release them and let the other country police them. The reason this might seem like an unrealistic scenario is not that POWs' claims for liberty are lost during times of war; it is that the conditions for holding paroled POWs accountable if they breach their duty have not existed and are unlikely ever to exist. (75)
In sum, the detention of POWs fits the AR model. Moreover, the AR model provides plausible normative guidance for the detention of STs. It holds that alien STs, whom the U.S. is not obliged to release and police, can be subjected to LTPD if they would be released to countries that would not adequately police them. At the same time, it holds that the U.S. must release and police its own citizen STs if it cannot convict them of a crime for which punitive detention, or loss of the right not to be subjected to LTPD, is a fitting punishment.
IV. THE UNSATISFYING NATURE OF MONITORING SUSPECTED TERRORISTS
If we proceed on the assumption that the U.S. may not subject its own citizens to LTPD as ECs, then it has to release and police those whom it cannot prove beyond a reasonable doubt have committed a serious crime that justifies long-term detention. One might be more or less sanguine about the prospects of such policing, however. If one is sanguine about it, then one might not feel the need to prosecute the ultimate inchoate crime, stating the intention to commit a crime. (76) One might prefer to wait until the suspected person has been caught attempting or conspiring to commit a crime.
The point of this Part of the Article is to argue that one should not be so sanguine about the powers of policing. Even if the police use the most sophisticated surveillance tools, augmented by some sort of restraining order that might be obtainable with evidence that does not rise to proof beyond a reasonable doubt of having committed a crime, (77) their policing will be relatively ineffective in preventing terrorist attacks. Of course surveillance and restraining orders would help ensure that STs do not successfully perpetrate a terrorist attack. My point is only that such tools have significantly limited effectiveness when compared with detention, and therefore, given the magnitude of the threat presented by terrorists, we have reason to prosecute the crime of threatening to commit a terrorist act.
Consider first what we can expect to do with surveillance. If there is probable cause to believe that a person is involved in criminal activity (including, of course, terrorist activity), a warrant could be obtained to wiretap his phone or computer communication. (78) Or, if there is probable cause to believe that he is an agent of an FTO, a FISA warrant can be obtained to do the same thing. (79) But someone plotting to commit a terrorist act, particularly once he has already been detained and questioned, must suspect that the government will be watching him in that way, and will avoid saying anything revealing on the phone or in mail (whether electronic or paper). He will communicate in person.
Another way to use surveillance would be to track his movements, either by putting a tracking device on his car (80) or even on his person. (81) But where one goes is not itself incriminating. If he went to the home of other STs, that might be helpful in building a conspiracy case. But it would only be a first step. Without more, there would be insufficient evidence to convict. (82)
The most useful form of surveillance would involve undercover agents who offer their services to help him carry out a terrorist plot. (83) If someone is not well networked into a terrorist organization that has done careful vetting, it would be hard for him to know that a person is an undercover agent who cannot be trusted. But if someone either has the capacity to operate essentially as a lone wolf, or has the right connections to help him vet his contacts, he will not easily be taken in by undercover agents.
A combination of restraining orders and monitoring could be more effective than monitoring alone. If someone is banned from buying guns or bomb-making materials, and then goes to a place where these things can be purchased, there would be reason to follow up and search his residence for breaking the restraining order. If he is banned from meeting with certain other individuals who are also STs, then he could be prosecuted for violating the order if he is seen in their company or at their homes. (84) The problem is that it would be impossible, without becoming a complete police state, to track all the people who might serve as intermediaries between STs. Thus, even this combination will have limited effectiveness against a cautious terrorist.
These limitations are illustrated by the occasional failure of the most draconian surveillance policies in the Western world--policies that go as close to preventive detention as possible without being preventive detention. The policies I am referring to are British "control orders." (85) There are two kinds of orders, the less restrictive ones that are held not to derogate from the European Convention on Human Rights, and the more restrictive ones that do derogate from that Convention. (86) Even the less restrictive orders--which can be obtained if the Secretary of State has "reasonable grounds for suspecting" that an individual is or has been involved in terrorist activities (87)--provide that those who are subjected to it may:
be electronically tagged at all times; be required to abide by curfews (by remaining at home or in a specified place between specified times, in the initial orders for up to 18 hours per day); have their passports taken away; be denied access to rail stations, airports, and ports; have their telephone lines cut; have their contacts with others outside the home severely limited; and be subject at any time to unannounced visits and searches from the police and security services. They may be prohibited from using specified articles or substances (for example mobile phones, faxes, pagers, computers, and internet facilities); have their right to work or engage in other activities limited; and suffer restrictions on association with other people, on communication, movement, and residence. They may be required to report to a specified person at specified times and places; ... to provide information on demand; and to permit anything to be removed from their home for the duration of the order.... The range of restrictions that may be imposed is without limit and may be renewed subject to twelve-monthly approvals by the courts. [Finally, b]reach of a condition imposed by a Control Order is a criminal offense punishable by imprisonment for up to five years. (88)
It is hard to imagine a regime for surveillance and restriction of behavior--other than house arrest or preventive detention in a secure facility--that would be more effective in preventing terrorist activity on the part of a particular individual. (89) Nonetheless, there are good reasons to doubt their efficacy.
Typically, controlees ... are allowed to arrange to attend group prayers at places of worship.... Controlees may be subject to curfew orders for large parts of the day and night but move freely within in preset areas outside those hours. It is questionable whether or not the determined terrorist will be unduly hindered by these patchy and inconsistent restrictions. As one suspect subject to a Control Order has observed: "I go everywhere now--on the underground, buses, the mosque. But I must be home by 7pm.... The government is playing games. If I am a security risk, why are they letting me out to be with people?" (90)
These points are confirmed by the fact that three of the first nineteen people subjected to control orders went missing. (91) Therefore, if the goal is to incapacitate people who have indicated their sincere intention to engage in terrorism, nothing short of detention--severely restricting a person's freedom to communicate with others and move about freely--will suffice.
V. THE CRIME OF MAKING TERRORIST THREATS
Threats paradigmatically involve two parties: those who make threats and those to whom the threats are directed. (92) The elements of the crime as traditionally understood include "that the threat actually caused the person threatened to be in sustained fear for his or her own safety or that of his or her immediate family." (93) There are, however, non-paradigmatic variations that lack this element and are nonetheless criminal. One way threats can lack this element is that they can fail to cause the relevant fear. Another way they can lack this element is if the actor has no reason to expect his victims to fear what he will do, but he has announced his intention to engage in violent, unlawful acts to others (although neither to his potential victims nor to people who he can reasonably expect will transmit his threat to his potential victims), thereby giving them reason to be concerned that he intends to commit a violent crime.
In the wake of Watts v. United States, the first Supreme Court case to discuss "true threats," (94) at least one court, the Fourth Circuit, was fairly sensitive to the significance of this last type of threat. (95) But as the doctrine evolved, confusion set in for both courts and commentators. In what follows, I explain first how the current doctrine of true threats is incoherent because of a failure to recognize that it encompasses two very different types of crimes: one dealing with causing fear or disruption (including both complete and inchoate crimes of causing fear and disruption), and the other an essentially inchoate crime directed at unlawfully and violently causing harm. Second, I explain how to make sense of the inchoate crime of stating the intention to commit unlawful, violent crimes, especially terrorism-while also developing the reasons for restricting the crime to threats to perform violent crimes. Third, I defend this inchoate crime against the objection that it takes the notion of inchoate crimes too far from completed criminal acts.
A. THE INCOHERENCE OF THE TRUE THREATS DOCTRINE
1. Proper Conceptual Framework for Threat Crimes
As a first step towards demonstrating the incoherence of threat doctrine, it is helpful to highlight the different interests the government has in preventing threats. These, as first spelled out by the Supreme Court in R.A.V. v. City of St. Paul, are: "protecting individuals from the fear of violence, from the disruption that fear engenders, and from the possibility that the threatened violence will occur." (96) The first two of these interests are importantly different from the third. The first two concern effects of threats directed at persons who feel threatened. Someone who threatens others and causes them to feel afraid or to disrupt their lives to avoid the threat has completed the harm by communicating his threat to his victims. The third is independent of the fear or disruption that might occur if potential victims know about the threat: it is an interest in protecting people from the threatened violence itself. With regard to that harm, the threat is just an indication that the person who makes the threat intends to cause that violence. The completed harm is not fear or disruption, it is the unlawfully- and violently-caused harm that is threatened. Threat law can help to prevent that harm if and only if it provides a basis for prosecuting people for stating their intention to perform unlawful, violent acts. In other words, it can prevent this third harm only by treating threats as a species of inchoate crime.
Of course, the first two harms can also ground inchoate crimes. An actor could not only culpably cause fear or disruption, he could also attempt to cause them, or recklessly or negligently act in a way that imposed too great a risk that they would be caused. The important divide, therefore, is not between the completed crime and the inchoate crime. It is between the effect that those crimes, if completed, would have: fear or disruption, or violently- and unlawfully-caused harm.
One might suggest that the state's interest in avoiding violently- and unlawfully-caused harm is not, by itself, a sufficient basis for the crime of making threats. That is, one might suggest that it is merely a supplemental interest the state has in preventing threats, paradigmatically understood. At least one case, however, United States v. Patillo, (97) treats this interest as one that can support a conviction for threats without the paradigmatic interests being implicated, and I argue below that this case was right to do so. Thus, the doctrine needs to be designed to accommodate this most inchoate of crimes.
To properly accommodate these different concerns, the law should distinguish four different threat crimes: three concerning fear and disruption, and one concerning the ultimately threatened harm. Each of these would have its own appropriate mens rea. All would involve knowingly making a statement and communicating it to others, but other mens rea standards would vary between the different crimes.
First, paradigmatic threats, which cause fear or disruption, should have a mens rea requirement of either knowledge that the statement would likely cause those harms, or a negligence standard that would make the statement culpable if a reasonable person would know that the statement would likely cause those harms. If one were particularly concerned with protecting free speech, one would have reason to favor the stronger knowledge requirement, or even require a specific intent to cause those harms; (98) if one were more concerned with preventing victims from feeling threatened, one would favor using a negligence standard instead. (99)
Second, attempted threats would cover those cases in which an actor intends to cause fear and disruption in his victim, but fails to do so. There are two ways an actor could fail to cause these harms. First, his victim could be impervious to the threat. Imagine a case in which A communicates what he thinks will be threatening to B, and B says: "Are you trying to threaten me? Well, you don't scare me." If A was trying to threaten B, but B is not scared, then A has attempted to threaten B. This is perfectly analogous to A trying to kill B by poisoning him, but failing to kill him because B is not susceptible to the poison that A used. Second, the failure could concern transmission of the threat. If the actor communicates the threat to a third party, or uses some medium like the mails, intending that it reach his ultimate victim, but the threat is not passed on or delivered, then the agent has attempted to threaten his victim. The mens rea in either kind of attempt would be, as with nearly all attempt crimes, (100) the specific intent to threaten. In other words, the mens rea would entail the actor intending that his threat be communicated to his victim and that his victim be scared as a result.
Third, negligent threatening would cover those cases in which an agent communicates a message that is objectively threatening to someone else, and expects or should expect it to reach his intended victim, causing fear or disruption, but it does not in fact reach his intended victim. (101) This is a crime akin to reckless or negligent endangerment, (102) the difference being that the harm is fear and disruption, not bodily harm. The mens rea requirement should be knowingly making a statement that a reasonable person would expect to both reach his intended victim and cause fear and disruption if and when it does so.
These first three crimes, being concerned with causing fear or disruption, should not require the actor to intend to follow through on the threat. The actor can cause these harms without intending to cause the ultimately threatened harm. (103) He can cause fear or disruption just as well by faking the intention to cause the ultimate harm as by actually having it.
The fourth crime, however, is different on this last point. It is concerned, not with causing fear or disruption in the intended victims of the threat, but with the ultimate harm associated with the threatened violence. To avoid making this a crime of merely being a danger to others--i.e., to retain some actus reus element--it is necessary to conceive of it as involving a choice to act. Therefore, it is necessary to understand this version of the crime as an inchoate crime: choosing to intend to commit the threatened harm and communicating that intention to others. In other words, this version of threat crime should be distinguished as the crime of stating the intention to unlawfully and violently harm others. And the mens rea for this crime, unlike the other three, has to include having the intention to cause the threatened harm.
To the best of my knowledge, no case or commentator has distinguished these four crimes before. As mentioned above, (104) at least one individual case shows some awareness of the need to make the fundamental distinction between threats that focus on fear and disruption and threats that focus on the ultimate harm. But others do not mark any of these distinctions and try to find one mens rea that fits them all. The result is an incoherent doctrine of true threats. I now survey a set of cases to illustrate both points.
2. Cases: Nuanced and Confused
The first case to frame the issue of true threats was Watts v. United States. (105) The case involved someone who said at a political rally protesting the draft: "If they ever make me carry a rifle the first man I want to get in my sights is L.B.J." (106) On that basis, he was tried for threatening the President. The Supreme Court held that his motion for acquittal at the end of the prosecution's case should have been granted because his statement, taken in context, was constitutionally protected political hyperbole, not a true threat. In the course of coming to that conclusion, the Court also expressed "grave doubts" that the statutory "willfulness" requirement required the defendant to intend to commit the crime. (107) Otherwise, it did not engage in any discussion of the required mens rea for a true threat.
The next important case to discuss true threats was Roy v. United States, which came up that same year. (108) The defendant in that case used a pay phone to tell the operator that he would kill the President if he came the next day, as he was planning to do, to the army base where the defendant was then stationed. The Ninth Circuit upheld his conviction for threatening the President, and interpreted the statutory "knowingly and willfully" mens rea element
to require only that the defendant intentionally make a statement, written or oral, in a context or under such circumstances wherein a reasonable person would foresee that the statement would be interpreted by those to whom the maker communicates the statement as a serious expression of an intention to inflict bodily harm upon or to take the life of the President, and that the statement not be the result of mistake, duress, or coercion. (109)
In other words, the court adopted a negligence standard of the sort that would be appropriate for a crime as grave as threatening the President. And taking a cue from the Supreme Court's grave doubts, the court rejected the thought that the speaker had to actually intend to carry out the threat, because that would "eviscerate the statute of its intended effect," (110) namely preventing people from making statements that "have a restrictive effect upon the free exercise of Presidential responsibilities." (111)
In adopting this negligence standard, the Roy court showed that it was concerned with preventing the harms associated with paradigmatic threats, in particular, having the President's schedule disrupted to avoid or otherwise deal with the threat. It failed, however, to distinguish completed threats, attempted threats, and negligent incomplete threats. And it failed to consider whether the prosecution should have to prove--given that this defendant's threat, like most threats to the President, did not actually disrupt his schedule--that the defendant had a specific intent to threaten. It simply assumed that the core crime is one that should be handled with a negligence standard, rather than a knowledge or specific intent standard. Nonetheless, recognizing that the purpose of the statute criminalizing threats to the President is primarily to prevent disruption of his activities, it properly rejected requiring that the defendant intend to carry out the threat.
Marking an important distinction from Roy, the Fourth Circuit, shortly afterward in its en banc decision in United States v. Patillo, (112) did require specific intent to harm the President. The defendant in that case made two comments to a fellow security guard at the Norfolk Naval Shipyard that he intended to kill the President. The court held
that where, as in Patillo's case, a true threat against the person of the President is uttered without communication to the President intended, the threat can form a basis for conviction under the terms of [18 U.S.C.] Section 871(a) [making it a federal crime to threaten the President] only if made with a present intention to do injury to the President. (113)
In other words, the court quite reasonably distinguished cases in which the threat could be expected to cause fear or disruption from cases in which it could not. In the former, the defendant need not intend to carry out the threat, but in the latter, the only government interest served by the statute is protecting the President, and that interest is triggered only if the defendant intends to carry out his threat. As the court noted, "[t]here is no danger to the President's safety from one who utters a threat and has no intent to actually do what he threatens." (114)
Despite the fact that these two cases--Patillo and Roy--were explicitly held to be consistent by the Fourth Circuit, (115) the Supreme Court agreed to hear Rogers v. United States to "resolve an apparent conflict among the Courts of Appeals concerning the elements of the offense proscribed by [18 U.S.C.] [section] 871(a)." (116) The "apparent" circuit split was between, on the one hand, the Roy court's use of an objective reasonableness standard, and, on the other hand, the Patillo court's use of a subjective intent standard. (117) The Court decided the case on other grounds, but Justice Marshall, joined by Justice Douglas, took the occasion to say that he would "interpret [section] 871 to require proof that the speaker intended his statement to be taken as a threat, even if he had no intention of actually carrying it out." (118) Justice Marshall's concern was in large part constitutional: the deterrence effect of the statute, on an objective reasonableness interpretation, would cause people to be too cautious about what they say, lest their words be interpreted as a threat even if they are not meant as one. This "would have substantial costs in discouraging the 'uninhibited, robust, and wide-open' debate that the First Amendment is intended to protect." (119) To avoid that effect, Marshall would have eliminated the crime of negligent, incomplete threats, reducing the number of threat crimes to two: complete threats and attempted threats, both with the same basic mens rea of specific intent to threaten.
It is noteworthy that the Rogers Court failed to recognize the reasonable distinction that the Fourth Circuit drew between Patillo and Roy. And while Justice Marshall's concern with protecting free speech is reasonable when it comes to paradigmatic threats, the essence of which is causing fear or disruption, he lost sight of the point the Fourth Circuit was making regarding threats that do "not involve the communication, or attempted communication, by a defendant of his threat to the President." (120) Moreover, he overlooked the possibility that a knowledge requirement-knowing or believing that the words used would be taken as a threat-would protect the First Amendment values he wanted to protect as well as the requirement of a specific intent to threaten. This failure reflects his failure to distinguish completed threats from attempted threats. And this failure to recognize basic distinctions was a sign of things to come. (121)
After the Rogers Court failed to embrace Justice Marshall's position on the specific intent to threaten, circuit courts generally used some version of an objective test for all threats, not just threats against the President. (122) The main split that evolved after Rogers was between different versions of objective standards. The Ninth Circuit in Roy used a reasonable speaker test, and many other circuits followed suit. (123) The Fourth Circuit, however, led a number of other courts in adopting the reasonable listener standard, according to which a case can be submitted to a jury if "there is substantial evidence that tends to show beyond a reasonable doubt that an ordinary, reasonable recipient who is familiar with the context of the letter (or message) would interpret it as a threat of injury." (124)
The Supreme Court again stepped into this narrow dispute in Virginia v. Black, (125) a case involving defendants who were prosecuted for violating Virginia's ban on burning a cross with the intent to intimidate. The Court held that "[i]ntimidation in the constitutionally proscribable sense of the word is a type of true threat, where a speaker directs a threat to a person or group of persons with the intent of placing the victim (or victims) in fear of bodily harm or death." (126) And it held that "[t]rue threats encompass those statements where the speaker means to communicate a serious expression of an intent to commit an act of unlawful violence to a particular individual or group of individuals." (127) The phrase "the speaker means to communicate a serious expression of an intent to commit an act of unlawful violence" suggests that the Court was adopting Justice Marshall's specific intent to threaten condition for true threats. (128) And thus, it might have seemed that the Court was trying to clarify the nature of "true threats," unprotected by the Constitution.
Unfortunately, the Court's decision has only added to a sense of confusion. First, the word "encompass" is ambiguous. Does it imply that only speech acts that involve the specific intent to threaten can be treated as true threats, or that those as well as some other acts constitute the full universe of true threats? Second, the case involved intimidation, which is a paradigmatic form of threat in which the victim is expected to experience fear. But did the Court mean to limit its discussion of the relevant mens rea to only those paradigmatic cases? Immediately after seemingly adopting Justice Marshall's specific intent-to-threaten test, the Court said that the "speaker need not actually intend to carry out the threat." (129) But then it quoted all three purposes for criminalizing threats mentioned in R.A.V., including "protecting people from the possibility that the threatened violence will occur." (130) Thus, one could take the Black Court's mens rea standard to apply to all threats, even mere statements of the intention to commit violent acts.
This confusion is both noted and replicated by the Seventh Circuit in United States v. Parr, (131) a case involving a defendant who was convicted of threatening to use a weapon of mass destruction against a federal government building, in violation of 18 U.S.C. [section] 2332a. First, with regard to what the Supreme Court meant in Black, the Parr court noted that "some circuits have held that a statement qualifies as a true threat only if the speaker subjectively intended it as a threat." (132) But it also noted that "[n]ot all courts have agreed that Black changed the test for true threats," (133) and that "[i]t is possible that the Court was not attempting a comprehensive redefinition of true threats in Black; the plurality's discussion of threat doctrine was very brief." (134) It then ducked the issue on the ground that it did not need to be resolved in Parr's case.
The confusion replicated in Parr concerns the failure to make a host of important distinctions. Parr communicated his threat to his cellmate in a Wisconsin prison, who then reported it to authorities. (135) There is no reason to believe that Parr wanted or expected his cellmate to communicate the threat to anyone else. Thus, this was not a paradigmatic threat in which the government had any interest in protecting people from fear or disruption. The government's only interest was in preventing the ultimate crime. In other words, the only criminal act Parr could have committed is that of stating his intention to commit a terrorist act. But the court did not take this as reason to require a specific intent to commit that ultimate act. Seemingly unaware of the decision in Patillo, the court simply cited Black for the proposition that "the government is not required to prove that the defendant in a threat case intended ... to carry out his threats." (136) In a similarly sweeping way, it held that a "threat doesn't need to be communicated directly to its victim...." (137) In so saying, it ignored the difference between communicating to a third party who would be expected to convey the threat to its victim and communicating to a third party who would not.
In sum, with Parr, we have come to a state in which the true threats doctrine is so muddled that courts no longer even recognize the distinctions that they are failing to make. While being uncertain about whether to use a specific intent-to-threaten standard, they have lost their sense that there are really four different crimes that count as true threats. Indeed, the courts have never paid attention to the differences between the first three crimes: completed paradigmatic threats, attempted paradigmatic threats, and recklessly or negligently communicating words that could reasonably be taken to be paradigmatic threats. Just by noticing those differences, much of the current confusion about the need for a specific intent-to-threaten test could be cleared up. But the difference that was once noticed between these paradigmatic threat crimes and the inchoate crime of stating the intention to commit unlawful, violent acts has now been lost. This is the more important difference, as this latter crime cannot reasonably require anything less than the intention to commit such violent acts. If the defendant had no such intention, then he cannot be charged with an inchoate crime connected to performing those acts.
In what follows, I leave to the side the dispute about the appropriate mens rea for paradigmatic threats and focus instead on defending the thought that there should be another kind of threat crime: the inchoate crime of stating the intention to commit unlawful, violent crimes.
B. MAKING SENSE OF THE INCHOATE CRIME OF STATING THE INTENTION TO COMMIT UNLAWFUL, VIOLENT CRIMES
Inchoate crimes are a controversial set of crimes. Many have been concerned that they really amount to punishing dangerousness, rather than wrongful choices. (138) The crime of stating the intention to commit unlawful, violent crimes, being the ultimate inchoate crime, makes those worries all the stronger. Nonetheless, I will argue here that those worries can be properly addressed. I do so in three steps. First, I spell out how this crime is like the classic inchoate crimes of attempt and conspiracy, which, when properly interpreted, do not merely penalize dangerousness but also wrongful choices. Second, I try to clarify when it makes sense to criminalize mere statements of intent--paradigrnatically when one is stating one's intention to perform a terrorist act that aims to cause death and destruction on a large scale, while being motivated by the kind of extremist ideology that motivates members of al Qaeda--and when the actor must have done more to acquire criminal liability. Third, I address some basic questions about the nature of the intention--how specific and how unconditional it must be--if the Government is to obtain a conviction. In the next section, I turn to responding to the objection that this crime is too far along the inchoate spectrum.
1. A Crime like Other Inchoate Crimes
The basic thought behind the crime of stating the intention to commit unlawful, violent acts, in particular terrorist acts, is that intentions are not mere thoughts; they are choices that one makes with regard to how one will behave. (139) Making such choices is the fundamental crime targeted by all notions of inchoate crime in which the person has not yet done anything that would constitute a completed crime. All inchoate crimes require some external act that provides evidence of the intention, (140) but statements can fill that role in the same way that a substantial step operates in attempt crimes and agreement operates in conspiracies.
This position on inchoate crimes presupposes the validity of the subjectivist model of criminal liability. This model, which fundamentally informs the Model Penal Code, (141) holds that the paradigmatic core of criminal liability is the choice to form, and act on, intentions that flout the law. (142) Obviously the set of restrictions on behavior imposed by the criminal law has to be concerned with more than mere intentions; it has to be concerned, most fundamentally, with preventing people from unjustly causing harms to others. But when it comes to explaining how an actor can be liable for committing a crime, the subjectivist focus on intentions makes more sense than an objectivist focus on causing harms. This becomes most clear when examining inchoate crimes.
The objectivist can say that an inchoate crime causes harm by "'disturbing the public repose,' ... or by causing apprehension, fear or alarm in the community because the actor has patently 'set out to do serious damage ... and to break the accepted rules of social life.'" (143) But the harm caused by any individual inchoate crime so pales in comparison to the harms of completed crimes, especially when the potential victim does not learn of the attempt, that the liability for these harms would, if harm is the basis for liability, be trivial. (144) If we want to make sense of the thought that inchoate crimes like attempted murder are serious crimes, even if the victim never knows about it, then we need to understand liability in a subjectivist way, as attaching to the criminal intent. This is not to deny that criminal effects are also relevant to the degree of culpability an actor has. (145) It is instead to say that the fundamental basis for culpability is the decision to pursue a criminal aim. Causing unjust harms if one completes a crime merely compounds that liability. (146)
If criminal intentions are the basis for criminal liability, then what is the relevance of an act in pursuit of them? In short, it is primarily epistemic: it confirms that the person has adopted the criminal intention. As William Blackstone put it:
For though, in foro conscientiae, a fixed design or will to do an unlawful act is almost as heinous as the commission of it, yet, as no temporal tribunal can search the heart, or fathom the intentions of the mind, otherwise than as they are demonstrated by outward actions, it therefore cannot punish for what it cannot know. (147)
Or, as George Fletcher put it, in the context of inchoate crimes, an act "is important [only] so far as it verifies the firmness of the [actor's] intent." (148) This verification has two components. On the one hand, people other than the actor cannot possibly know his intentions until he has manifested them in some way, in action or in speech. On the other hand, equally important if often overlooked, it is necessary to distinguish mere thoughts, and in particular fantasies about how one might want to act, from sincerely adopted intentions. (149)
The law may not criminalize mere fantasies. To allow freedom of conscience, (150) the state has to allow people to make whatever value judgments they want to make about any acts--even criminal acts. They can fantasize about committing whatever crimes appeal to them. They are limited only in what they decide to do. This crucial distinction between values and fantasies on the one side, and decisions regarding actions on the other, is marked by the choice to form an intention. Yet it is not always easy to distinguish intentions from mere fantasies. Indeed, it may be hard for an actor to be certain whether he has really formed an intention to do something criminal until he has in some way started to act on the intention. Thus, the act requirement serves not only to provide external evidence of what an actor intends, it also serves to distinguish a sincere and firm intention from a fantasy.
Of course, a single act or statement could reflect many things other than a sincere or firm intention to commit a criminal act. In the attempt context, one might engage in acts preparatory to a crime, such as buying a gun, only as a way of indulging a fantasy with no intention of carrying out the crime. One might even go further towards the completion of a crime just to experience a frisson of excitement as the completion comes near, again without any intention of completing the crime. Likewise, in the conspiracy context, one might make statements indicating that one intends to partake in a criminal enterprise without sincerely intending to do so, effectively making a false promise to the other members of the conspiracy. And one might make a statement of one's intention to commit a crime, outside the context of an agreement that is not a true threat but is instead a joke of some sort, or an expression of bravado or rage.
All of these possibilities raise the real worry that a conviction for an inchoate crime is a false positive--a conviction for adopting a criminal intention when the actor has done no such thing. But false positives are always a worry in criminal law. All one can demand of acts is that they corroborate having a sincere and firm intention. (151) The rest of the work necessary to establish that a person has indeed formed a sincere and firm intention has to come from the context in which the acts are performed and the background character of the actor. If the question is whether an actor's substantial step towards the completion of a crime was merely a form of play, then knowing whether he engages in such play regularly is relevant. And so is knowing what he has said to others, and what possible motives he may have had in doing so. These are difficult issues, but they are handled on a regular basis in prosecution of attempt crimes and conspiracies. (152) Relying on a statement as opposed to a substantial step or an agreement raises no new difficulty. (153)
One might object that I misrepresent the essence of inchoate crimes as the formation of a criminal intention. Joshua Dressier, for example, describes the formation of the intention as simply the mens rea of inchoate crimes, saying that there is no crime until there is also an actus reus. His reason for saying this is that the alternative is to punish someone for "thoughts alone." (154) He makes the point in the context of describing a six-stage sequence for the commission of a completed crime:
First, the actor conceives the idea of committing a crime. Second, she evaluates the idea, in order to determine if she should proceed. Third, she fully forms the intention to go forward. Fourth, she prepares to commit the crime, for example, by obtaining any instruments necessary for its commission. Fifth, she commences commission of the offense. Sixth, she completes her actions, thereby achieving her immediate criminal goal. (155)
This is a useful model for the commission of a completed crime, and Dressier is surely fight that attempt liability attaches only at stage five. But he would be wrong if he were to suggest (he never says this explicitly) that this is the stage at which there first arises a potential actus reus. Clearly there is action even at stage four. But more to the point, there is choice to engage in action at stage three. Punishment for mere thought--the having and entertaining of ideas, the weighing of reasons, the experiencing of emotions, sensations, and perceptions--would occur only if people were punished for their mental activity in stages one or two.
To be clear, I am not saying that the law can criminalize any and every form of choice. One can choose, by directing one's attention one way or another, whether to have certain kinds of thoughts as well as whether to take actions. The freedom of thought must include the freedom to choose to indulge any thoughts one wants to indulge. My point is that the law can criminalize choosing to form an intention to pursue a criminal plan of action without criminalizing mere thought. If the completed action would be criminal, the law can criminalize the choice to take such an action, and that choice is what one makes when one forms the intention to do so. (156)
It is also important to be clear that I am not suggesting anything that comes close to the model of policing represented in the Minority Report, (157) that of predicting criminality and punishing on the basis of that prediction. What I am calling for is criminalizing a choice to engage in criminal conduct. Of course if such choices had no bearing on the commission of criminal acts, then it would be pointless to criminalize them. But there is an intrinsic connection between forming an intention and going on to perform the intended actions; intentions function to guide us. (158) The connection is not perfect; intentions can always be revised or renounced. (159) Even so, one can have no legitimate reason to set oneself to the task of performing criminal acts. One might have legitimate reason to pretend to perform criminal acts, or to intend to perform many of the preparatory acts that would normally lead to a criminal act--one might, for example, want to understand, for constructive purposes, just how easy it is to be in position to complete a criminal act. But assuming the criminal law is just and that one has a duty to obey just laws, (160) then one should not intend to complete a criminal act.
Importantly, what matters in an individual case is not whether, having formed such an intention, one is so likely to perform the act that one has to be preventively stopped. That is, the concern with intentions is not simply a concern with a more or less reliable mechanism for making predictions of later criminality. What matters in forming a criminal intention is that one has already chosen to flout the law. Any punishment would then be for that criminal act, performed in the past.
If the choice to intend to do something criminal is the core of any inchoate offense, then one might ask: Why criminalize only statements at a stage earlier than Dressler's fifth stage of commencing the commission of the crime? In other words, if statements can be a sufficiently reliable indicator of a criminal intention after stage three--the adoption of the criminal intention--why not also punish on the basis of other, nonverbal behavior, such as preparatory action at stage four, which is indicative of having formed the intention to commit a crime?
I can give no satisfying categorical answer, but I can point out that nonverbal actions that do not fall into category five are very likely to be equivocal. Without an accompanying statement to show that they are in pursuit of a criminal intention, it is hard to judge that someone has, beyond a reasonable doubt, started to act on a criminal intention. (161) Indeed, the line between stages four and five is inherently vague. What really marks the line is that acts in stage four can easily be given a benign interpretation, whereas acts in stage five more often cannot. (Consider, for example, the difference between buying a gun and tracking down an intended victim while carrying the gun.) Statements, which may be offered well before the actor's actions reach stage five, may clarify an actor's intention in ways that mere behavior cannot.
2. Breadth of the Crime of Stating a Criminal Intent
The logic of this subpart up to this point might seem to justify prosecuting actors for stating their intention to engage in any criminal actions, not just terrorism. But there is a balance to be struck between, on the one hand, respect for freedom and avoidance of an overly intrusive police state, and, on the other hand, the need to prevent seriously harmful acts that would otherwise be hard to prevent. This balance calls for limiting the crime of stating one's criminal intent to a narrow range of completed crimes for which the harm is great and deterrence is unlikely to work; crimes like terrorism.
The reason not to make this crime too broad is grounded in the basic importance of liberty, including the liberty to do immoral but not harmful things. This is not just a point about the state not taking sides in reasonable disagreements about religious ideals or other conceptions of a good life. (162) It is a point about the state recognizing that some wrongs are personal in a way that makes it inappropriate for the state to get involved. Because of this respect for individual liberty, the state should generally be reluctant to punish. It should punish only when the need for doing so, in terms of preventing unjust harms and upholding the integrity law, is high. (163) Flouting the law is therefore a necessary but not a sufficient reason for punishment, (164) again assuming the law is just. (165)
To illustrate the point that not all wrongs should be covered by the criminal law, consider marital infidelity. Quite often such acts inflict real harms; nevertheless, the personal intimacies that would have to be explored to decide what sort of punishment an unfaithful spouse deserves are the kinds of concerns that do not fit well into a prosecutor's portfolio, at least, not in a liberal society. This is not to say that courts have no role in adjudicating marital matters, but their role should be limited to tasks like facilitating an equitable divorce, if matters come to that point, and ensuring that children are not abused, rather than punishing one party for failing to live up to her marital commitments. Likewise, statements of criminal intentions ought normally to be left in the private sphere. Again, this is not to deny that the underlying intentions are morally wrong. It is to say that they are far enough from leading to harm, and are part of what is normally considered the private realm of life--planning how to live--that the proper rebuke for publicly adopting a criminal intention should normally be social, not criminal. (166)
One might think that another reason not to criminalize statements of the intent to commit a crime is that it becomes harder to prove that someone really would commit a completed crime the further he is from committing it. As Robert Chesney put the point:
The farther that one moves from the paradigm of a completed act--as one moves backwards successively through attempt, to advanced planning, to initial planning, and so forth--the more tenuous the link between the defendant and the anticipated harm becomes and, hence, the more likely it is that false positives will be generated. (167)
There are, however, two reasons why this is not a good reason to limit the crime. First, it presupposes the objectivist position rejected above--that the only real crime is causing an unjust harm. Indeed, Chesney's way of putting things, in terms of a link between the defendant and the anticipated harm, makes it seem that prosecutions for inchoate crimes are really just attempts to predict who will commit a crime. If that were what inchoate crime prosecutions were about, they would be a variation on the theme from Minority Report. (168) The state cannot justly punish for a crime that has not been committed. (169) If the fundamental crime is forming a criminal intent, however, then there can be no false positives if the agent has indeed formed such an intention. Of course, he might not go on to commit the completed crime, but that does not mean that he has not already acted wrongly.
Second, there may, of course, be false positives with regard to the claim that an actor has formed a criminal intention, but that does not mean that the risk of false convictions has to be unusually high. Even if it is harder to be certain that an actor has formed a criminal intention the more inchoate his crime, that means only that the supporting evidence has to be that much stronger. It does not mean that we ought to allow the standard of proof beyond a reasonable doubt to be relaxed the further the inchoate crime is from a completed crime. And if the standard of proof remains the same, then there is no reason to think that there will be more false positives the more inchoate a crime is.
The reason to limit the crime of stating the intention to commit a crime, then, is that forming intentions is presumptively part of the private sphere of life that the state should not regulate with the criminal law. What weighs on the other side, in exceptional cases, is the seriousness of the harm and the inability of the threat of criminal punishment to deter certain potential criminals from committing their intended crimes.
Crimes are at their most serious when they threaten a whole community, such as genocide, or when they inflict large-scale death and destruction. They are still very serious when they victimize only a few or even one person, but inflict the most serious harms on those persons: permanent and severe harm (physical or mental) or death. Crimes are essentially impossible to deter when the would-be perpetrator would rather accept the punishment, or even death, than fail to inflict the harm he wants to inflict. (170) Such crimes arise when the motive is not economic gain but some overwhelming desire, such as an unyielding thirst for revenge or extremist convictions, like those found in some with extremist religious or political beliefs.
Given these two positive factors, stating one's intention to perform a terrorist act that aims to cause death and destruction on a large scale, while being motivated by the kind of extremist ideology that motivates members of al Qaeda, would be the paradigmatic criminal act of stating one's intention to commit a further crime. Other complete crimes that might fit this mold would be political crimes like assassination of leading government figures, especially the President. And given that death is intended in other cases, one could at least argue for extending the crime to threats against the lives of non-public individuals, at least when the person making the threat cares more about causing harm than about any possible punishment--as all too often seems to be the case when a jealous and vengeful domestic or former domestic partner decides that he would rather be dead or in jail than see his partner or former partner move on with her life. (171)
It is worth noting that this question of breadth could also be asked of other inchoate crimes. Why should all attempted crimes count as crimes? Perhaps it is easy enough to see why all complete but imperfect attempts--crimes in which the actor has performed the last act needed to complete the crime, but in some way he missed his target and failed to cause the harm he intended to cause--should be criminal. The actor has not only formed a criminal intent, he has followed through on it as far as he could. The same cannot be said of an actor who has taken a substantial step (Dressler's stage five), but has not completed his attempt. Perhaps he would decide not to follow through. (172)
It seems to me that the answer can be found in the thought that there is a continuum of liability from first forming the intention, to starting to act on it, to completing the plan, to causing the harm. At various crucial junctures, such as starting to prepare (Dressler's stage four) and taking substantial steps (Dressler's stage five), the actor should reconsider the intention. Failure to do so compounds his culpability and gives the criminal law more reason to take an interest in his actions. By stage five, the actor is close enough to a completed crime that his claim to be operating in a private space is outweighed by the state's interest in upholding the integrity of the law. The really interesting cases should be stage four cases. As noted above, without a statement clarifying what the preparatory actions mean, they are too equivocal for grounding a prosecution. (173) But the combination of a statement of intentions and preparatory actions should arguably justify criminal prosecutions in a wider range of cases than just a statement or an action alone. How wide a range I leave for another occasion.
3. Nature of the Intention: How Specific and Unconditional
The main concern for the specificity of an intention is that if it is sufficiently vague, then there is reason to doubt that it has been sincerely and firmly adopted. In other words, vague, open-ended plans to commit terrorist acts are more likely than well-developed plans to be idle fantasies rather than sincere intentions. It is, of course, impossible to be precise about just how well developed a plan must be to count as a sincere intention. But the Parr case provides some seemingly reasonable guidance on what it takes to act on a "plan [that] was detailed": (174)
He said he planned to construct a bomb inside a delivery truck, park the truck outside the federal building, and walk inside as if to make a delivery. He explained that he would briefly enter the building but then would slip outside immediately and run as far as he could before the bomb exploded. (175)
He also "mentioned the number of detonators and drums of explosives he would use, where he would park, and how he would deflect suspicion." (176) The court notes that there were some large undefined parts of his plan. For example, he "did not specify an exact time frame for his plan, noting that he would be on probation for eight years after his release and would use that time to 'refine [his] techniques' because he would get only one chance. But he vowed to pull off his plan within ten years." (177) Altogether, given that "Parr emphasized that he 'absolutely' would execute his plan," (178) the court had no trouble concluding that Parr sincerely intended to communicate a threat. It did not address whether this would be a sufficiently well-defined intention to count as an intention to carry out the threat--it mistakenly thought that he did not have to intend to carry out the threat. (179) But the portrait the court paints is of someone whose intention was sufficiently specific to count as a sincere intention to commit a terrorist act.
Conditionality allows for some more clear-cut distinctions, but also presents difficult line-drawing cases. One general distinction in the criminal law dealing with crimes which require the mens rea of specific intention is that a conditional intention counts as the full specific intention "unless the condition negatives the harm or evil sought to be prevented by the law defining the offense." (180) On that view, the intent to blow up a building unless the government accedes to one's wishes--a conditional intention because the government might accede to one's wishes, in which case one would not blow up the building--counts the same as the unconditional intention to blow up the building. The condition, far from negating the evil, trades on it. By contrast, if one intended to blow something up unless one discovers that it is against the law to do so, that would not count as the intention to blow something up in a way that violates the law.
More problematically, what should we say about someone who intends to blow up a building if he gets the chance? By the formula just described, he should be treated as if he has the unqualified intention to blow up the building. But now suppose that the building is in a heavily guarded military base, and he believes he will have a chance to blow it up only if (a) he is invited onto the base and (b) the policy of searching people who enter the base for guns and explosives is discontinued. It seems quite unlikely that he would ever have a chance, given those conditions, to act on his intention, and we can suppose that he knows just how unlikely it is. Can it still be reasonable to say that he should be treated as if he has an unconditional intention to blow up that building? That seems unreasonable, but how should we draw the relevant line?
I suggest the following gets at the heart of the core distinction between intention and fantasy: if the condition is sufficiently likely to arise that the actor is actively or regularly watching to see if it has arisen, and if he is ready to act on his illicit intention if the condition does arise, then he should be treated as if he has the unconditional intention. By actively or regularly monitoring conditions to determine whether he can at some point act on it, and by otherwise being prepared to act on it, he is actively engaging it. If, by contrast, the condition is sufficiently unlikely that he has no need to keep himself in a state of ready watchfulness, then the intention should be treated as mere conditional intention. In that case, even if he has behaved improperly by adopting that conditional intention, he is not sufficiently engaged by it to be criminally culpable. If and when the condition arises, he will then have to choose whether to act on the intention. He might well recant at that time, when he reengages actively with the intention. Until that time, then, the preference should be for liberty over punishment.
For an illustration, imagine someone who announces that he will attack U.S. government buildings if the U.S. ever invades Yemen. This is a live possibility at this time, in mid-2011. But what is his attitude towards the intention? Does he do anything to prepare himself to attack U.S. government buildings in anticipation of this possible eventuality? Or does he just go about his life, waiting passively for news of such an invasion, without doing anything to prepare for it? If the latter applies, then his threat should not be taken as a fully formed intention to commit a terrorist act; if the former applies, then it should. (181)
C. OBJECTION AND REPLY
In this last subpart, I will entertain the objection that it is improper to prosecute people for merely forming the intention to commit terrorist acts. The arguments I address were formulated by Larry Alexander and Kim Ferzan. (182) They start with the same assumption that I accept, namely that forming an intention is an act in its own right that is potentially culpable. (183) But they go on to offer four reasons to conclude that it is not the kind of culpable act with which the law should be concerned. I respond here to all four reasons in turn.
1. Distinguishing Intentions from Desires and Fantasies
Alexander and Ferzan argue that it is difficult to distinguish intentions from desires or fantasies, which are not acts but states of mind. (184) I have already discussed this problem above, (185) but I highlight here the way in which stating the intention to commit unlawful, violent acts compares with other inchoate crimes in this regard. Consider, first, how attempt law addresses this problem. Acting on an intention in a way that takes it nearly to completion is a good sign that one has adopted an intention, as opposed to merely feeling a desire or playing with a fantasy. Now consider conspiracy law. Having to forge an agreement with another focuses the mind on just what one is really willing to do, (186) and thus making an agreement is another way of testing the sincerity of an intention, and helping to distinguish it from a mere desire or fantasy. Mere statements by themselves are a less reliable sign that someone has sincerely adopted an intention. But the sincerity with which the message is communicated, whether it is repeated, the context in which it is said, and the person's general character all suffice to establish, at least in some cases, that the intention was sincerely formed. (187)
2. The Conditionality of Intentions and the Opacity of Future Circumstances
Alexander and Ferzan's second argument concerns the conditionality of intentions and the opacity of future circumstances. I have just discussed the issue of conditionality. (188) But Alexander and Ferzan raise a number of finer objections, that I have not yet considered, having to do with conditions that the actor may not yet have considered. (189) One such objection concerns the possibility that an intention that seemed unproblematic would be problematic if certain unanticipated conditions arise. Their example concerns someone who is intending to pick up her child from school in ten minutes. (190) Given the traffic conditions at the start of the drive, she should be able to do so without driving recklessly. But if she hits a traffic jam, she may have to speed to get there on time. What is her level of culpability when she first forms the intention?
I suggest that the answer should depend on whether she has anticipated this possibility and formed the conditional intention to act impermissibly if it does arise. If she has formed that conditional intention, then she has formed an illicit intention. If the harm is serious enough, then acting on this conditional intention should be criminally culpable. If she has not anticipated this possibility, however, then the mere disposition to behave impermissibly cannot be the basis of criminal liability. She has not yet made a choice to behave impermissibly.
Another problematic example concerns someone who decides to blow up a football stadium, but has not decided whether to do so when it is full of people or empty. (191) The magnitude of the threatened crime varies tremendously depending on whether it is just property or also lives that she threatens. Given that she has not addressed that issue, how culpable is her intention?
I suggest that the answer should depend on whether there is reason to believe that, when putting her plan into action, she would kill innocents. If she shows a general indifference to the lives of others, or sees some extra value in killing innocents, then her culpability should be high, perhaps nearly as high as if she intended to kill innocents. But if she generally shows respect for human life, and it can be assumed that she would later refine her plan to blow up the stadium when people are not in it, then her culpability should be low, as it would be if she intended only to cause property damage. What if there is no way to tell how she would develop the intention? One way that might be true is if the prosecution has no knowledge of her character and larger motives. If that is the situation, the case should probably be dropped because there will be too little evidence to assert that she sincerely intended anything. The other way it may be impossible to know her intention is if the prosecutor is aware of competing factors that might well be in equipoise to the actor who had not yet sorted them out herself. In that case, splitting the difference between the high and low levels of culpability seems a reasonable solution, one somewhat akin to prosecuting someone for actions with reckless indifference to human welfare.
It is worth pointing out that this problem is one that arises not just for the crime of stating the intent to unlawfully and violently harm others. It also applies to attempted crimes and conspiracies in which it is unclear how certain details would be worked out when the crime is interrupted and the defendant arrested. Just as this problem has not proven fatal for those inchoate crimes, so, too, should it not be fatal to the crime of stating one's criminal intent.
3. Duration and Renunciation of Intentions
Alexander and Ferzan next argue that the proposition that culpability attaches to the formation of an intention gets caught on a dilemma having to do with duration and culpability. 192 On the one hand, one might think that duration--the length of time an actor holds an intention--is not relevant to culpability, because culpability attaches when one forms a criminal intention. This is supported by the idea, argued for above, (193) that culpability depends on having made the wrongful choice to form the criminal intention. It is also supported by the thought that the risk to the potential victim does not depend on how long one has formed an intention, as long as one has sufficient time to act on it.
On the other hand, this position seems to lead to implausible conclusions at the extremes. Alexander and Ferzan imagine a case in which Hank has decided on January 1 of one year to kill Sally on January 1 of the next year. Six months later, Harry forms the same intention to kill Sally on the next January 1. Suppose that Harry "revokes his intention after thirty seconds, whereas Hank has held his intention firmly for more than six months [and then revokes it]." (194) The implication of the supposition is clearly that duration must matter, that Harry cannot be as culpable as Hank. And then, to push the point further, they suppose that Harry vacillates between forming the intention and then revoking it. Each choice to form it must be a new culpable choice, making him guilty of many culpable acts, in contrast to Hank's one culpable act. This, they note, is implausible: "it seems odd to deem indecisive Harry more culpable and deserving of more punishment than steadfast Hank." (195)
I think the right reply to this purported dilemma is as follows. First, with regard to the second problem, that of vacillation, note that for each completed crime that one might intend to perform, there can be only one intention to perform it (with many variations regarding how to perform it). Hank may form it and hold onto it, and Harry may form it, drop it, accept it again, and so on. But what Harry is doing is vacillating with regard to a single intention. He does not commit a second culpable act when he accepts it again; he reenters the stream of his earlier culpability, losing whatever benefit may have come to him from renunciation. With regard to the first problem, it seems problematic only because thirty seconds is so brief that one doubts that Harry sincerely held the intention. Even if Harry changed his mind because of new information about Sally, we think that one has to live with an intention for a while before one can say that it is firmly held. It is hard to say just how long such a while is, but if we move from seconds to weeks, then it is not so counter-intuitive to say that Hank and Harry are equally culpable if Hank has had the intention to kill Sally for six months and Harry for two weeks. Of course, if one acts on an intention right away, and completes the crime, then one is no longer in a position to renounce one's intention. (196) But if one has not acted on it, or has only started to act on it, then one can continue to ponder whether one really sincerely embraces it. And it normally takes some time before such pondering can result in a conclusion that is settled and firm.
Finally, Alexander and Ferzan argue that it does not make sense to hold people culpable for forming criminal intentions, because until they perform the final act necessary to cause harm, they can and should revoke their intentions, and would be non-culpable, at least in certain jurisdictions, (197) if they did revoke their intentions--something which is not true of any other completed crime that one tries to "undo." (198) Their argument seems to be: (a) that once an actor revokes his criminal intention, he should not be culpable and (b) that the only reason to think otherwise is the false belief that by forming a criminal intention, he imposes a risk on his potential victims that he will carry it out. They do not argue in support of the first part of this argument, but they do argue in support of the second. It is a false belief, they say, because forming a criminal intention is not like "lighting ... a long fuse." (199) The latter necessarily raises the risk to potential victims because the actor might not be able to cut the fuse before it reaches the dynamite. The former is different because a rational actor always has the capacity to change his mind about an intention and always has reason--the reason that explains why harming another is wrong in the first place--to do so.
The problems with their argument are multiple. One can reject both (a) and (b), and the substantive position that there is something inherently suspect about having a defense that allows people to undo only certain crimes. I start with (b). Alexander and Ferzan overlook the idea that one might reject the defense of abandonment or revocation, not because one thinks that when a person adopts a criminal intention he raises the odds of causing the ultimate harm, but rather because one thinks that when a person adopts a criminal intention, he has wrongly chosen to flout the law. (200) If one holds that view, then one can take one of two positions on (a): either revocation should exists as an anomalous complete defense, or it should not be a complete defense at all. I personally think both positions are plausible. But it is sufficient for my purposes here to show that at least one of them is plausible. Accordingly, I finish this subpart by defending the idea that abandonment of a criminal intention should not be a complete defense to a crime, but rather should function analogously with the partial defense available to someone who has performed the last act, and then tried to undo or prevent the harm he has unleashed. That is, it should limit one's culpability, but it should not altogether erase it.
How can abandonment or renunciation even limit one's culpability if one's culpability is established by having formed a sincere and firm intention to commit a criminal act? The answer is that at various crucial junctures, when one acts on the criminal intention, one's culpability grows because one has shown a firm commitment to committing the crime in the face of more concrete reasons to abandon the intention. Thus, one's culpability increases as one moves from mere intention formation, to starting to prepare, to commencing the commitment of the crime itself, to completing the last act needed to perform to complete the crime. At every step along the way, one can renounce the criminal intention. If one renounces out of principle, one has done something punishment seeks to cause one to do, which is to recognize the error of one's intention. That should limit one's culpability, and perhaps even earn one some forgiveness. (201)
In other words, there is a coherent view according to which culpability can grow, not merely due to the passage of time during which one holds a criminal intention, but due to actions taken to bring the intention to fruition. This alternative view fits perfectly well with the commonly shared intuition that renunciation of a criminal intention can limit or reduce culpability, and that going further and further towards the completed crime can raise culpability. And it also fits the view that culpability starts to set in when one has sincerely formed a criminal intention--the act requirement being necessary as an evidentiary matter, but not as a fundamental moral matter.
Finally, it is worth pointing out again, as I did after their second argument, that this fourth argument applies equally against attempt law, at least the kind the follows the Model Penal Code's suggestion that culpability attaches after one has taken a substantial step towards completion of the crime. For Alexander and Ferzan, that, too, cannot be defended in the face of the possibility of renunciation. (202) But I would turn that around on them and say that just as one can be culpable for an attempted crime after taking a substantial step towards completing it--despite the fact that one could still change one's mind--so can one be culpable for forming the intention to commit a violent crime, and announcing it to others, despite the fact that one could still change one's mind. (203) Particularly when dealing with the threat to commit terrorist acts, with their potentially devastating results, there is good reason to want the criminal law to step in and prevent the act from occurring as soon as a culpable act based on that intention has been performed--that is, as soon as the person has sincerely stated the intention.
The autonomy respecting model of detention implies that those who are functionally autonomous adult citizens of a country may not be preventively detained for a long time, so long as the state's normal policing powers are operative. This conclusion may seem hard to stomach when it comes to certain hard cases: people who announce that they are bent on committing terrorist acts. What society can let such people free among them? But it may seem that there are no good alternatives. If these people cannot be prosecuted for crimes such as conspiring or attempting to commit terrorist acts (or ancillary crimes), then they must be either preventively detained or released and policed. The first of these options is indefensible in a liberal society; and the latter seems to provide inadequate security. I have argued here, however, that the law of threatening allows for a third option. They can be prosecuted for the ultimate inchoate crime: stating the intention to commit unlawful, violent acts.
This is not legally controversial, but it is nonetheless philosophically problematic. The doctrine concerning threat law is a mess, and has failed to clearly distinguish the crime that concerns causing fear and disruption from the crime that concerns stating that one has the intention to commit a particular violent crime. But the distinction can be made and defended. Forming the intention to commit a criminal act is the essence of inchoate crimes. And while the crime of stating the intention to commit unlawful, violent acts pushes the outer limits of the idea of an inchoate crime, it does not surpass those limits. As long as the crime itself is sufficiently serious, and the prospects for deterrence sufficiently low, there is reason to have such a crime. Further, those conditions are met when dealing with politically or religiously motivated terrorist crime.
To make use of the crime of stating one's intention to commit a terrorist act, the state would have to prove beyond a reasonable doubt that the person was expressing a real intention, a true threat of the non paradigmatic sort, rather than demonstrating his bravado, expressing his fantasies, or venting his rage. That would not be an easy case to make. But given evidence about what was said, how it was said, character, motive, and context, it should be possible. And note, the same challenge would confront military authorities trying to decide whether to continue to detain a particular detainee who expressed his desire to commit terrorist acts, as some detainees at Guantanamo have done. (204) The burden of proof is higher for a criminal case, but that is as it should be. The state should not be in the business of detaining its own autonomous citizens simply because it thinks they are dangerous, as long as the state's normal policing capacity is intact. It should subject them to long-term detention only if it can convict them of a serious crime. The crime of stating the intention to commit terrorist acts is just that.
(1) For example, 18 U.S.C. [section] 2332b(a)(2) makes it a crime, punishable by up to ten years in prison, to threaten to commit an act of terrorism transcending national boundaries. 18 U.S.C. [section] 2332b(a)(2) (2006).
(2) See Corpus Juris Secundum on Threats, 86 C.J.S. Threats [section] 3 (2010). It lists, as one of the elements of a criminal threat, "that the threat conveyed to the person threatened a gravity of purpose and the immediate prospect of execution of the threat." Id. In other words, a threat paradigmatically involves a threatened person.
(3) This is directly contrary to the view taken by Paul T. Crane, "True Threats" and the Issue of Intent, 92 VA. L. REV. 1225, 1270 (2006).
(4) This doctrine traces back to Watts v. United States, 394 U.S. 705 (1969).
(5) Alec Walen, A Unified Theory of Detention, with Application to Preventive Detention for Suspected Terrorists, 70 MD. L. REV. (forthcoming 2011).
(6) A FISA warrant can be obtained if a federal officer shows probable cause to believe that the target is an agent of a foreign power, including organizations such as al Qaeda. 50 U.S.C. [section][section] 1801(b)(1)(C); 1801(c); 1805(a)(2)(A) (2006).
(7) The jurisdictional bases for prosecuting for an act of terrorism transcending national boundaries are very broad, and it would be hard to blow up a major building in a city like Los Angeles without meeting them. See 18 U.S.C. [section] 2332b(b)(1) (2006). The same is true for 18 U.S.C. [section] 2332a (concerning the use of weapons of mass destruction; also including a threat crime). See 18 U.S.C. [section] 2332a (2006).
(8) A could not currently be tried by a military commission, because those are reserved for the trial of aliens. Military Commissions Act (MCA), 10 U.S.C. [section] 948c (2006). But suppose the MCA was revised to allow U.S. citizens to be tried by military commission. This presumably could constitutionally be done, based on Ex parte Quirin, 317 U.S. 1 (1942) (holding that U.S. citizens who worked for an enemy nation could be tried in a military commission for war crimes), and al-Marri v. Pucciarelli, 534 F.3d 213 (4th Cir. 2008) (en banc), vacated and remanded sub nora. al-Marri v. Spagone, 129 S.Ct. 1545 (2009) (holding that the Authorization for the Use of Military Force (AUMF), Pub. L. No. 107-40, 115 Stat. 224 (2001), allows the President to treat al Qaeda as an enemy like equivalent to an enemy nation). And it would help the Government bring in U's testimony without making U available for cross-examination. The MCA states that hearsay evidence can be introduced if "direct testimony from the witness is not available as a practical matter, taking into consideration the physical location of the witness, the unique circumstances of military and intelligence operations during hostilities, and the adverse impacts on military or intelligence operations that would likely result from the production of the witness." 10 U.S.C. [section] 949a(b)(3)(D)(ii)(III) (2006). But [section] 949a(b)(3)(D)(ii)(IV) would allow U's testimony to be introduced as hearsay, not allowing cross-examination, only if "the interests of justice will best be served by admission of the statement into evidence." Id. [section] 949a(b)(3)(D)(ii)(IV). Because U's testimony is the heart of the Government's case, it is hard to believe that this last prong would be met.
(9) 18 U.S.C. [section] 2339B (2006).
(10) See, e.g., RICHARD B. ZABEL & JAMES J. BENJAMIN, JR., HUMAN RIGHTS FIRST, IN PURSUIT OF JUSTICE: PROSECUTING TERRORISM CASES IN THE FEDERAL COURTS (2008), available at http://www.humanrightsfirst.org/wp-content/uploads/pdf/080521-USLS- pursuitjustice.pdf.
(11) These are only the most obvious crimes the government could seek to use to prosecute A before he causes any harm, For a more complete list, see ZABEL & BENJAMIN, supra note 10, ch. V.
(12) See, e.g., al-Marri, 534 F.3d at 284-93, 293-351 (opinions of Judges Williams and Wilkinson); Jack Goldsmith, Don't Try Terrorists, Lock Them Up, N.Y. TIMES, Oct. 9, 2010, at A21.
(13) See President Barack Obama, Protecting Our Security and Our Values (May 21, 2009) (transcript available at the National Archives) [hereinafter Protecting Our Security]; see also Periodic Review of Individuals Detained at Guantanamo Bay Naval Station Pursuant to the Authorization for the Use of Military Force, Exec. Order No. 13,567, 76 Fed. Reg. 13,277 (Mar. 7, 2011), available at http://www.whitehouse.gov/the-press-office/2011/03/07/ executive-order-periodic-review-individuals-detained-guant-namo-bay-nava [hereinafter Periodic Review] (more recent executive order establishing procedures for the ongoing military detention of Guantanamo detainees).
(14) Protecting Our Security, supra note 13.
(15) See Hamdi v. Rumsfeld, 542 U.S. 507 (2004) (O'Connor, J., plurality opinion) (discussing the need for adequate procedural protections for U.S. citizens held as enemy combatants).
(16) See id. at 533-34; see also Periodic Review, supra note 13 (detailing new procedures for review of combatant status); Robert Chesney & Jack Goldsmith, Terrorism and the Convergence of Criminal and Military Detention Models, 60 STAN. L. REV. 1079 (2008) (pointing out that the procedures for enemy combatants are evolving towards those of criminal trials, but not all the way).
(17) See Walen, supra note 5, at subpart III.B.
(18) See U.S. v. Salerno, 481 U.S. 739 (1987) (upholding the Bail Reform Act of 1984).
(19) See Mathew Waxman, Administrative Detention of Terrorists: Why Detain, and Detain Whom?, 3 J. NAT'L SECURITY L. & POL'Y 1, 35 (2009).
(20) See Zadvydas v. Davis, 533 U.S. 678, 701 (2001); Walen, supra note 5 (discussing the Geneva Convention Relative to Protection of Civilian Persons in Time of War, Aug. 12, 1949, 6 U.S.T. 3516, 75 U.N.T.S. 287, art. 78); see also Speedy Trial Act, 18 U.S.C. [section] 3161(c)(1) (2006)).
(21) I focus only on U.S. citizens because even legally resident aliens can be deported for suspected terrorist activities. Insofar as they are prosecuted, they have the same due process rights as U.S. citizens. But they lack the same fundamental right to be released and policed if not prosecuted. See Walen, supra note 5, at subpart III.E.
(22) I use the concept of autonomy here in the descriptive sense identified by Richard H. Fallon, Jr. in Two Senses of Autonomy, 46 STAN. L. REV. 875 (1994).
(23) This sort of connection between conviction and detention is also embraced, for constitutional reasons, by Justices Scalia and Stevens in Hamdi v. Rumsfeld, 542 U.S. 507, 554 (2004) (Scalia, J., dissenting).
(24) This example is modeled off the case of Steven Parr, who told his cellmate in a Wisconsin prison that he intended to blow up the Reuss Federal Plaza in Milwaukee. The cellmate then sent a letter warning the FBI of Parr's threats. Parr was convicted of threatening to use a weapon of mass destruction against a federal building in violation of 18 U.S.C. [section] 2332a(a)(3) (2006). See United States v. Parr, 545 F.3d 491 (7th Cir. 2008).
(25) People do say such things to their captors. Consider the case of Abdallah al-Ajmi, a Kuwaiti held in Guantanamo from 2002 to 2005. An officer at an administrative review board hearing testified: "In August of 2004, Al Ajmi wanted to make sure that ... the tribunal members know that he is now a jihadist, an enemy combatant and that he would kill as many Americans as he possible [sic] can." Rajiv Chandrasekaran, From Captive to Suicide Bomber, WASH. POST, Feb. 22, 2009, at A1.
(26) Stephen J. Morse says almost exactly this about armed robbers: "Imagine ... a three time convicted armed robber who threatens, completely believably, to commit a fourth crime.... In the legal world we now inhabit ... criminal conviction in the absence of at least attempted crime would [not] be possible to prevent future dangerous conduct." Stephen J. Morse, Neither Desert nor Disease, 5 LEGAL THEORY 265, 265-66 (1999).
(27) See supra note 7 (discussing 18 U.S.C [section][section] 2332a and 2332b (2006)). It is worth adding that the Convention on the Suppression of Unlawful Acts Related to International Civil Aviation, Art.1(3), also makes it a crime under international law to "credibly threaten" various unlawful acts against international civil aviation. International Civil Aviation Organization, Convention on the Suppression of Unlawful Acts Related to International Civil Aviation, art.1(3) (Sept. 10, 2010), available at http://www.icao.int/DCAS2010/restr/ docs/beijing_convention_multi.pdf.
(28) The maximum sentence under [section] 2332b for threatening to commit an act of international terrorism is ten years, but the maximum sentence under [section] 2332a for threatening to use a weapon of mass destruction is life in prison. In addition, the U.S. Sentencing Guidelines allow a twelve-level enhancement for "an offense ... that involved or was intended to promote, a federal crime of terrorism." U.S. SENTENCING GUIDELINES MANUAL [section] 3A1.4 (2002).
(29) As Justice Breyer put it: "Freedom from imprisonment--from government custody, detention, or other forms of physical restraint--lies at the heart of the liberty that [the Due Process] Clause protects." Zadvydas v. Davis, 533 U.S. 678, 690 (2001).
(30) Reasons other than that a witness is undercover could include that the evidence was tainted by torture, or cruel, inhuman, or degrading treatment; that the evidence was gathered in a war zone by military officers who were not concerned with preserving it for prosecutorial use; or that the evidence may include sensitive intelligence sources and methods. But see GUANTANAMO REVIEW TASK FORCE, FINAL REPORT 23 (2010), available at http://www.justice.gov/ag/guantanamo-review-final_report.pdf [hereinafter FINAL REPORT] (stating that the "principal obstacles to prosecution in the cases [of Guantanamo detainees] deemed infeasible ... typically did not stem from concerns over protecting sensitive sources or methods from disclosure, or concerns that the evidence against the detainee was tainted").
(31) Walen, supra note 5.
(32) See Michael L. Corrado, Sex Offenders, Unlawful Combatants, and Preventive Detention, 84 N.C.L. REV. 77 (2005); Morse, supra note 26; Christopher Slobogin, A Jurisprudence of Dangerousness, 98 Nw. U. L. REV. 1 (2003).
(33) Corrado's argument has the ring of a principled argument, but it is fundamentally pragmatic, turning on questions of line-drawing and abuse of power. Corrado, supra note 32, at 108-09. Morse takes a position that starts with an assumption very much in line with my autonomy-respecting position: that "[t]he state must leave people alone ... unless they are nonresponsibly dangerous or unless they harm others." Morse, supra note 26, at 294. But then he shows that he is willing to balance that assumption with the thought that "[e]veryone also has a right to be free from unjustifiable, intentional harm from others." Id. Thus, insofar as we can predict that someone will be dangerous--something he thinks we cannot do now and are unlikely to be able to do in the foreseeable future--we can use pure preventive detention on them. Slobogin's argument is more thoroughly principled, but it is also unsound, claiming that those who cannot be deterred are outside the reach of the criminal law, and that it is only those who are outside the reach of the criminal law that can be subject to LTPD. Slobogin, supra note 32, at 5. This is an unsound argument for two basic reasons. First, it is simply false that those who cannot be deterred because they are more dedicated to committing a crime than to avoiding punishment are outside the reach of the criminal law. As responsible agents, they may fully deserve the punishment they get. Second, it is also not true that only those who may not be subjected to criminal punishment may be subjected to LTPD, as the case of prisoners of war (POWs) who commit war crimes demonstrates.
(34) See U.S. CONST. art. I, [section] 9, cl. 2 (allowing the suspension of the writ of habeas corpus "when in Cases of Rebellion or Invasion the public Safety may require it").
(35) 542 U.S. 507, 519 (2004). My position is a moral one, but following Ronald Dworkin I think that constitutional principles, such as the protection of liberty by the Due Process Clauses of the Fifth and Fourteenth Amendments, should be interpreted by appeal to the underlying moral principles, at least insofar as doing so is consistent with the text and history of constitutional interpretation. See RONALD DWORKIN, FREEDOM'S LAW: THE MORAL READING OF THE AMERICAN CONSTITUTION 7-11 (1996).
(36) Justice Souter objected that the Non-Detention Act bars the government from holding its own citizens as ECs. Hamdi, 542 U.S. at 541. But this objection is statutory, not constitutional.
(37) 317 U.S. 1 (1942).
(38) 542 U.S. at 519. The plurality opinion left it unclear whether U.S. citizens could be detained only if fighting for a traditional national enemy on a traditional battlefield, or if they could also be detained if they fought for al Qaeda, with no affiliation to a government or former government of a nation that, like the Taliban, attacked the U.S. Id. at 521 ("If the practical circumstances of a given conflict are entirely unlike those of the conflicts that informed the development of the law of war, that understanding may unravel. But that is not the situation we face as of this date. Active combat operations against Taliban fighters apparently are ongoing in Afghanistan."); see also Corrado, supra note 32, at 117-18. Other courts have divided on that point. For example, the Fourth Circuit in al-Marri voted 5-to-4 to allow legal resident aliens, and by implication, U.S. citizens--an implication that the court accepted but that I reject--to be detained as ECs. See al-Marri v. Pucciarelli, 534 F.3d 213 (4th Cir. 2008) (en banc), vacated and remanded sub nora. al-Marri v. Spagone, 129 S. Ct. 1545 (2009).
(39) See Hamdi, 542 U.S. at 554 (Scalia, J., dissenting).
(40) The Court takes this position in Exparte Quirin. 317 U.S. at 37-38. Some scholars criticize Quirin on this basis. See, e.g., David J. Danelski, The Saboteur's Case, 1 J. SUP. CT. HIST. 61, 79 (1996) (arguing that the Court blurred the line between the law of war and military law, only the latter of which is enacted by Congress).
(41) Being unaccountable is not the same as being undeterrable. Contra Slobogin, supra note 32, at 4.
(42) See Pub. Comm. Against Torture in Israel v. Gov't of Israel, H.C.J. 769/02, [section] 61 (Dec. 11, 2005), available at http://elyon1.court.gov.il/files_eng/02/690/007/a34/ 02007690.a34.pdf.
(43) Id. (internal citations omitted).
(44) Thus the International Committee of the Red Cross, the official reporter of the Geneva Conventions, defines international humanitarian law, also known as the law of war, as "a set of rules which seek, for humanitarian reasons, to limit the effects of armed conflict." See INT'L COMM. OF THE RED CROSS, WHAT IS INTERNATIONAL LAW? (2004), available at http://www.icrc.org/eng/resources/documents/legal-fact-sheet/humanitarian- law-factsheet.htm.
(45) See, e.g., JEFF MCMAHAN, KILLING IN WAR (2009); MICHAEL WALZER, JUST AND UNJUST WARTS (2d ed. 1992); Christopher Kutz, The Difference Uniforms Make: Collective Violence in Criminal Law and War, 33 PHIL. & PUB. AFF. 148 (2005).
(46) See F.M. Kamm, Non-Consequentialism, the Person as an End-in-Itself and the Significance of Status, 21 PHIL. & PUB. AVE. 354, 382-83 (1992).
(47) Consider the sacrifice jurors are required to make for the sake of providing justice to the community.
(48) For example, the mentally ill who are a danger to themselves or others. See Addington v. Texas, 441 U.S. 418, 426 (1979) (noting the state's "power to protect the community from the dangerous tendencies of some who are mentally ill").
(49) This is an idea that I develop in Alec Walen, A Punitive Precondition for Preventive Detention: Lost Status as an Element of a Just Punishment, SAN DIEGO L. REV. (forthcoming 2011). I argue that this view provides the best possible account of a number of detention practices, such as the civil (preventive) detention of sexually violent predators who have served their prison sentences. See Kansas v. Hendricks, 521 U.S. 346 (1997).
(50) The obvious example here is quarantine. See Compagnie Francaise de Navigation a Vapeur v. La. State Bd. of Health, 186 U.S. 380, 387 (1902) ("[S]tate quarantine laws and state laws for the purpose of preventing, eradicating, or controlling the spread of contagious or infectious diseases, are not repugnant to the Constitution....").
(51) See Atkins v. Virginia, 536 U.S. 304, 318 (2002) (discussing the diminished culpability of the mentally retarded). They may still be held accountable, but their disability mitigates their potential culpability, and makes it inappropriate to hold them fully accountable. Id.
(52) See Nathaniel Berman, Privileging Combat? Contemporary Conflict and the Legal Construction of War, 43 COLUM. J. TRANSNAT'L L. 1, 9 (2004).
(53) See Geneva Convention Relative to the Treatment of Prisoners of War art. 91, Aug. 12, 1949, 6 U.S.T. 3316, 75 U.N.T.S. 135 [hereinafter GC III] ("Prisoners of war who have made good their escape [including by rejoining their own armed forces] and who are recaptured, shall not be liable to any punishment in respect of their previous escape"--implying that escaped prisoners have the right to take up arms again).
(54) See THE HANDBOOK OF HUMANITARIAN LAW IN ARMED CONFLICT 361 (Dieter Fleck ed., 1995).
(55) This assumption seems sound as a general matter in immigration law, but there might be exceptions for cases in which the U.S. is responsible for depriving a detainee of the opportunity to make a decent life for himself elsewhere. That is arguably the case for the Uighur detainees held in Guantanamo. But see Kiyemba v. Obama, 555 F.3d 1022 (D.C. Cir. 2009) (holding the federal courts lack the authority to order Guantanamo detainees released into the United States).
(56) See FINAL REPORT, supra note 30, at 18 (discussing the security problems in Yemen and the way they interfere with releasing Yemeni detainees from Guantanamo).
(57) In this regard, it makes sense that roughly 40% of the detainees left in Guantanamo are from Yemen. See id. at 14.
(58) As Justice Souter wrote, explaining his interpretation of the AUMF: "[T]here is no reason to think Congress might have perceived any need to augment Executive power to deal with dangerous citizens within the United States, given the well-stocked statutory arsenal of defined criminal offenses covering the gamut of actions that a citizen sympathetic to terrorists might commit." Hamdi v. Rumsfeld, 542 U.S. 507, 547 (2004) (Souter, J, concurring in part, dissenting in part); see also id. at 554 (Scalia, J., dissenting) ("No one contends that the congressional Authorization for Use of Military Force, on which the Government relies to justify its actions here, is an implementation of the Suspension Clause [suspending the writ of habeas corpus].").
(59) See Scott Shane, Year of Plots: A 2nd Prism, N.Y. TIMES, Jan. 13, 2010, at A1 ("Exactly 14 of the approximately 14,000 murders in the United States last year resulted from allegedly jibadist attacks: 13 people shot at Fort Hood in Texas in November and one at a military recruiting station in Little Rock, Ark., in June.").
(60) See, e.g., Scott Shane & Eric Lipton, Passengers' Quick Action Halted Attack, N.Y. TIMES, Dec. 26, 2009, at A1 (describing how if Umar Farouk Abdulmutallab had been a bit more competent, he would have killed hundreds on his Detroit-bound Northwest Airlines flight on Christmas Day 2009).
(61) See David Luban, Unthinking the Ticking Bomb (Georgetown Law Faculty Working Papers, Paper 68, 2008), available at http://scholarship.law.georgetown.edu/fwps_papers/68.
(62) See Christopher C. Joyner & Alexander Ian Parkhouse, Nuclear Terrorism in a Globalizing World: Assessing the Threat and the Emerging Management Regime, 45 STAN. J. INT'L L. 203,205 n.4 (discussing a wide range of estimates, from 1% to 50%, that experts give of the likelihood of a terrorist nuclear attack in the next decade).
(63) My conclusion here is very much like one that David Cole reaches when he says:
Given conspiracy laws, it is difficult to imagine cases where the government has reliable evidence that an individual is going to commit an imminent terrorist act, but lacks probable cause of any criminal activity. If it lacks even probable cause, society should take the risk associated with continued surveillance, as we do with all other crimes, rather than permit preventive detention.
David Cole, Out of the Shadows: Preventive Detention, Suspected Terrorists, and War, 97 CALIF. L. REV. 693, 747 (2009). The primary difference is that I think one has to add the crime of making terrorist threats to conspiracy law to close the gap between security needs and criminal law's tools.
(64) Hamdi v. Rumsfeld, 542 U.S. 507, 518 (2004) ("The capture and detention of lawful combatants and the capture, detention, and trial of unlawful combatants, by universal agreement and practice, are important incidents of war.") (internal quotation marks and square brackets omitted).
(65) GC III, supra note 53, at art. 21; see also Berman, supra note 52, at 9 n.14 ("This practice may now seem somewhat quaint, but it effectively highlights the purpose of the detention of prisoners of war.").
(66) JEAN DE PREUX, THE GENEVA CONVENTIONS OF 12 AUGUST 1949: COMMENTARY 181 (Jean Pictet ed., 1958) [hereinafter COMMENTARY GC III] ("[A] prisoner of war who is released on parole and is recaptured bearing arms may be tried and sentenced by the Detaining Power.").
(67) GC III, supra note 53, at art. 21, [paragraph] 2. Parole seems to have been used in World War II for those who might need to be released from a detention facility for reasons of "health or hygiene." COMMENTARY GC III, supra note 65, at 179. The obvious reason it was restricted to these conditions is that a broader usage would have been too dangerous for the detaining powers.
(68) GC III, supra note 53, at art. 21, [paragraph] 2: "Prisoners of war may be partially or wholly released on parole or promise, in so far as is allowed by the laws of the Power on which they depend." (emphasis added). In other words, the countries for which they fight may not allow them to give parole.
(69) Code of Conduct for Members of the Armed Forces of the United States, Exec. Order No. 10,631, 20 Fed. Reg. 6057, 3 C.F.R. 1954-58 Comp. 266 (1955), as amended by Exec. Order No. 12,017, 42 Fed. Reg. 57941 (1977); Exec. Order No. 12,633, 53 Fed. Reg. 10355 (1988). The Department of Defense also says, "The United States does not authorize any Military Service member to sign or enter into any such parole agreement." Dep't of Defense Directive 1300.7, Enclosure 2, [paragraph] B3a(5). This policy makes sense insofar as soldiers might seek to use capture and parole as a way of evading their military duty.
(70) GC III, supra note 53, at art. 21, [paragraph] 3 (emphasis added).
(71) COMMENTARY GC III, supra note 65, at 182.
(72) "A person who gives his parole gives a personal undertaking on his honour for which he is in the first place responsible to himself." Id.
(73) See supra notes 55-56 and accompanying text.
(74) I am not saying that parole is a pointless piece of dead letter law. It was sometimes used in World War II, and one can imagine that it would be used again in certain limited cases, such as when a detainee, due to ill health, seems very unlikely to take up arms again.
(75) One might want to object that the law of war was not designed to respect liberal notions of autonomy. In fact, it was negotiated over generations between countries, many of which were in no way liberal. As its alternative name, humanitarian law, suggests, it was designed simply to minimize the brutality of war while also allowing the warring parties to take those steps essential to their security. See In re Yamashita, 327 U.S. 1, 15 (1946) (describing the purpose of the law of war as being "to protect civilian populations and prisoners of war from brutality"). Military necessity, one might say, provided the only reason for not releasing POWs before the cessation of active hostilities. In response, I am happy to admit that the AR model was not in play in devising the law of war. Nonetheless, I think it is an important moral fact that the LTPD of POWs is consistent with respecting the liberty of autonomous agents. That is why we, as citizens of a liberal country, should be willing to accept the practice going forward.
(76) I refer to stating the intention to commit terrorist acts as the ultimate inchoate crime because it attaches criminal liability to nothing but forming a criminal intention and expressing that fact. Arguably conspiracy law at least sometimes operates the same way, when it requires no overt act beyond an agreement. See the MODEL PENAL CODE [section] 5.03(5) (1985) (not requiring an overt act for conspiracies to commit first- or second-degree felonies). But even then, the crime of stating a terrorist threat is at the far end of the inchoate spectrum.
(77) A restraining order or civil protection order, at least in the context of domestic violence, typically can be obtained on a temporary or emergency basis ex parte. A "permanent" order, that typically lasts one to two years, will be awarded only after a hearing in which the normal rules of civil procedure apply. An order will then be granted only if the facts alleged in the petition are supported by a preponderance of the evidence, including evidence that the threatening behavior is likely to continue. See Judith A. Smith, Battered Non-Wives and Unequal Protection-Order Coverage: A Call for Reform, 23 YALE L. & POL'Y REV. 93, 100-01 (2005).
(78) U.S. CONST. amend IV.
(79) See supra note 6.
(80) This was recently upheld as not being a search under the Fourth Amendment in United States v. Pineda-Moreno, 591 F.3d 1212 (9th Cir. 2010), petition for cert. filed, No. 10-7515 (U.S. Nov. 10, 2010).
(81) Clearly, one does have an expectation of privacy in one's person, even if not in one's car. Therefore some sort of probable cause would be required to get a warrant before such a tracking device could be used. See Kyllo v. United States, 533 U.S. 27, 33 (2001) ("[A] Fourth Amendment search occurs when the government violates a subjective expectation of privacy that society recognizes as reasonable.").
(82) Conspiracy presupposes an agreement. See MODEL PENAL CODE [section] 5.03(1) (1985). Merely meeting with people is clearly insufficient to establish agreement.
(83) See, e.g., Colin Miner, Liz Robbins & Erik Eckholm, Bomb Plot Foiled at Holiday Event in Portland, Ore., N.Y. TIMES, Nov. 28, 2010, at A1 (describing the case of Mohamed Osman Mohamud, a nineteen-year-old naturalized U.S. citizen from Somalia, who was arrested after he tried to blow up a fake bomb in the center of Portland, Oregon, where the bomb was given to him by an undercover agent).
(84) Bloom v. Illinois, 391 U.S. 194, 201 (1968) ("Criminal contempt is a crime in the ordinary sense.").
(85) These were established by the Prevention of Terrorism Act, 2005, c. 2 (U.K.) [hereinafter PTA].
(86) See Lucia Zedner, Preventive Justice or Pre-Punishment? The Case of Control Orders, 60 CURRENT LEGAL PROBS. 174, 176-78 (2007).
(87) Id. at 176 (citing PTA, supra note 85, at [section] 2(1)(a)).
(88) Id. at 179-80.
(89) Indeed, this sort of regime seems extreme enough to be of dubious constitutionality. The British House of Lords ruled in 2008 that eighteen-hour curfews amounted to a deprivation of the fight to liberty, as protected by the UK's Human Rights Act. Afua Hirsh, 16-Hour Curfews Violate Rights, Court Hears, GUARDIAN (U.K.), May 6, 2010, at 19. Insofar as the U.S. Constitution embraces similar standards as the UK's Human Rights Act, a similar policy in the U.S. could run into constitutional problems. However, it could also be argued that if the U.S. can detain STs in LTPD, then it can impose on them a less severe deprivation of their liberty, such as is found in control orders.
(90) Zedner, supra note 86, at 190-91.
(91) Alan Travis & Alex Kumi, Manhunt as Terror Suspect Escapes Control Order, GUARDIAN (U.K.), Jan. 17, 2007, at 6.
(92) See Corpus Juris Secondum on Threats, supra note 2, at [section] 3.
(94) See supra note 4.
(95) See United States v. Patillo, 431 F.2d 293 (4th Cir. 1970) (en banc), aff'd on reh'g, 438 F.2d 13 (4th Cir. 1971) (en banc).
(96) 505 U.S. 377, 388 (1992). The Court offered these reasons in the context of illustrating the thought that it is acceptable to criminalize a subset of unprotected speech (threats) because of a particularly pronounced concern with the reason that category of speech is unprotected. One of the examples the Court used to make that point involved criminalizing threats to the President, saying that "the reasons why threats of violence are outside the First Amendment ... have special force when applied to the person of the President." Id.
(97) 431 F.2d 293 (4th Cir. 1970) (en banc), aff'd on reh'g, 438 F.2d 13 (4th Cir. 1971) (en banc).
(98) As I explain in greater detail in the next subsection, Justices Marshall and Douglas, in their concurring opinion in Rogers v. United States, 422 U.S. 35, 48 (1975), take the stronger position, "that the speaker intended his statement to be taken as a threat," i.e., that he had the specific intent to threaten.
(99) I have in mind a double negligence standard, looking to both reasonable speakers and reasonable listeners. Most courts that adopt a negligence standard impose a reasonableness standard on only one side. See Crane, supra note 3, at 1243-48 (discussing reasonable speaker, reasonable listener, and "reasonable neutral" tests).
(100) See JOSHUA DRESSLER, UNDERSTANDING CRIMINAL LAW 391 (5th ed. 2009). But see COLO. REV. STAT. [section] 18-2-101(1) (2010) (linking the mens rea of attempt with that of the completed crime); see also Reilly v. State, 55 P.3d 1259 (Wyo. 2002); Brown v. State, 790 So. 2d 389 (Fla. 2001); People v. Thomas, 729 P.2d 972 (Colo. 1986) (all allowing convictions for attempt based on recklessness).
(101) I assume that this crime, being an inchoate crime would merge with the completed crime, as all inchoate crimes except for conspiracy do. See SANFORD H. KADISH, STEPHEN J. SCHULHOFER & CAROL S. STEIKER, CRIMINAL LAW AND ITS PROCESSES: CASES AND MATERIALS 674 (8th ed 2007).
(102) See MODEL PENAL CODE [section] 211.2 (1985) (Recklessly Endangering Another Person).
(103) Justice Marshall correctly notes this in Rogers, 422 U.S. at 47 ("[D]efendant is subject to prosecution for any statement that might reasonably be interpreted as a threat, regardless of the speaker's intention.").
(104) See supra text accompanying note 95.
(105) 394 U.S. 705 (1969).
(106) Id. at 706.
(107) Id. at 708.
(108) 416 F.2d 874 (9th Cir. 1969).
(109) Id. at 877-78.
(110) Id. at 878 n.15.
(111) Id. at 877.
(112) 431 F.2d 293 (4th Cir. 1970) (en banc).
(113) Id. at 297-98.
(114) Id. at 298.
(115) The Fourth Circuit reconsidered and then affirmed its prior decision in United States v. Patillo, 438 F.2d 13 (4th Cir. 1971) (en banc). In doing so, the court explicitly held that its holding was consistent with Roy. Id at 15.
(116) 422 U.S. 35, 36 (1975).
(117) See id. at 43 n.1 (listing cases and circuits).
(118) Id at 48.
(119) Id. (quoting N.Y. Times Co. v. Sullivan, 376 U. S. 254, 270 (1964)).
(120) United States v. Patillo, 438 F.2d 293 (4th Cir. 1970) (en banc).
(121) This failure to recognize the different kinds of threats and the different, arguably reasonable mens rea requirements is also reflected in the secondary literature. See, e.g., G. Robert Blakey & Brian J. Murray, Threats, Free Speech, and the Jurisprudence of the Federal Criminal Law, 2002 BYU L. REV. 829 (2002); Crane, supra note 3, at 1239.
(122) See Crane, supra note 3, at 1243 nn.76-77 (collecting cases). One exceptional case was United States v. Twine, 853 F.2d 676 (9th Cir. 1988) (requiting specific intent to threaten when threatening someone other than the President).
(123) See Crane, supra note 3, at 1244-46 (collecting cases).
(124) United States v. Maisonet, 484 F.2d 1356, 1358 (4th Cir. 1973). Other circuits that followed this lead, including the Second, Seventh, Eighth, and Eleventh Circuits. See Crane, supra note 3, at 1246 nn.96-99 (collecting cases). The difference between the tests is that in the reasonable listener tests the objective standard becomes a jurisdictional test (whether the case can go to the jury), and the State has to prove only one mens rea element: knowingly making the statement. See Crane, supra note 3, at 1246-47. It is not clear what difference this would make in practice.
(125) 538 U.S. 343 (2003).
(126) Id. at 360.
(127) Id. at 359 (internal quotation marks omitted).
(128) See Rogers v. United States, 422 U.S. 35, 48 (1975) (Marshall, J., concurring).
(129) Black, 538 U.S. at 360.
(130) Id. (internal quotation marks omitted).
(131) 545 F.3d 491 (7th Cir. 2008).
(132) Id. at 499.
(133) Id. at 499 n.2.
(134) Id. at 500.
(135) Id. at 494.
(136) Id. at 498.
(137) Id. at 497. I am aware of only one other case, one outside the federal courts, which upheld a conviction for threatening when it was the defendant did not intend the threat to be communicated to a potential victim. Beard v. United States, 535 A.2d 1373, 1377-78 (D.C. 1988) (upholding a conviction for threatening a juror despite the contention that the defendant did not intend for that juror to know of the threat). This case too did not examine the significance of making a non-paradigmatic threat, but it backed away from a clear holding that a crime is committed if no one feels threatened, ending the discussion of the communication of a threat by noting that the defendant "ha[d] reason to believe the hearer [of his threat, another juror] will be upset or intimidated." Id. at 1378.
(138) See, e.g., LARRY ALEXANDER & KIMBERLY KESSLER FERZAN (WITH STEPHEN MORSE), CRIME AND CULPABILITY: A THEORY OF CRIMINAL LAW 199 (2009); Morse, supra note 26, at 287-89; see also Robert M. Chesney, Beyond Conspiracy? Anticipatory Prosecution and the Challenge of Unaffiliated Terrorism, 80 S. CALIF. L. REV. 425, 434-35 (2007).
(139) See generally MICHAEL BRATMAN, INTENTION, PLANS, AND PRACTICAL REASON (1987) (describing a planning conception of intentions); see also ALEXANDER & KESSLER FERZAN, supra note 138, at 200-01; Alec Walen, The Doctrine of Illicit Intentions, 34 PHIL. & PUB. AFF. 39, 47 (2006).
(140) This is true, at least, of the classic inchoate crimes: attempt, conspiracy, and solicitation. It is arguably not true of another class of crimes that are like these inchoate crimes in being anticipatory, that is, in not involving a harm in themselves but in aiming to prevent harms: possession crimes. See Douglas Husak, Rethinking the Act Requirement, 28 CARDOZO L. REV. 2437 (2007).
(141) See DRESSLER, supra note 100, at 385.
(142) Dressier mistakenly represents the core concern of subjectivists as being "dangerousness and bad character." Id. These are not choices, and they may not be the basis for punishment.
(143) See DRESSLER, supra note 100, at 386 (quoting Clark v. State, 8 S.W. 145, 147 (Tenn. 1888); Thomas Weigend, Why Lady Eldon Should Be Acquitted: The Social Harm in Attempting the Impossible, 27 DEPAUL L. REV. 231, 264 (1977)).
(144) The crime for which Parr was convicted carried a ten-year sentence--hardly trivial--and that was after the judge sentenced him to a below-Guidelines term. United States v. Parr, 545 F.3d 491, 496 (7th Cir. 2008).
(145) ALEXANDER & FERZAN, supra note 138, at ch. 5, do deny that effects are relevant to the degree of culpability. I reject their arguments for this conclusion in Alec Walen, Crime, Culpability and Moral Luck, 29 LAW & PHIL. 373 (2010).
(146) See Walen, supra note 145 (defending the last claim that actual harm compounds liability).
(147) 4 WILLIAM BLACKSTONE, COMMENTARIES *21.
(148) GEORGE FLETCHER, RETHINKING CRIMINAL LAW 138 (1978).
(149) One might trace this concern with distinguishing real intents from fantasies back to Gerald Dworkin & David Blumenfeld, Punishment for Intentions, 75 MIND 396, 401 (1966).
(150) See Cantwell v. Connecticut, 310 U.S. 296, 303 (1939) (stating that the First Amendment freedom of religion includes freedom of conscience).
(151) But see the Model Penal Code on attempt, requiring that the substantial steps be "strongly corroborative of the actor's criminal purpose." MODEL PENAL CODE [section] 5.01(2) (1985).
(152) See, e.g., People v. Lauria, 251 Cal. App. 2d 471 (1967) (discussing many reasons why intent might or might not be found in a conspiracy case). But see the critical discussion of this case in KADISH, SCHULHOFER & STEIKER, supra note 101, at 700.
(153) See GLANVILLE LLEWELYN WILLIAMS, CRIMINAL LAW: THE GENERAL PART 630 (2d ed. 1961) (discussing the ambiguity of acts unaccompanied by statements, and the more reliable information, for the purposes of establishing mens tea, provided by a confession of a guilty intention).
(154) DRESSLER, supra note 100, at 379.
(156) Connecting this with the AR model discussed in Part III, one can be punitively detained on that model if one has autonomously chosen to commit a crime. The choice to frame and act on an intention to engage in a crime is just such a choice.
(157) MINORITY REPORT (Twentieth Century Fox Film Corp. 2002).
(158) See BRATMAN, supra note 139, at 27 (stating that intentions "are conduct-controlling pro-attitudes, they have inertia, and they serve as inputs into further practical reasoning").
(159) Id. at 24 (stating that "unless and until I do give up or reconsider my prior intention, its role in my means-ends reasoning will be to set an end for that reasoning,..." implying that one can give up or reconsider one's intentions).
(160) This is of course a controversial assertion. It may be true for crimes that are malum in se. But for crimes that are malum prohibitum, or even for crimes that some believe are malum in se but which others think should not be criminal act at all, the basis for having a duty to avoid committing those crimes would have to turn on a political account of the duty to respect legitimate law. For one influential view of that, see JOHN RAWLS, THE IDEA OF PUBLIC REASON REVISITED, reprinted in THE LAW OF PEOPLES 137 (1999) ("[L]egitimate law ... may not be thought the most reasonable, or the most appropriate, by each, but it is politically (morally) binding on him or her as a citizen and is to be accepted as such."). For a friendly critique of this position, see Alec Walen, Reasonable Illegal Force: Justice and Legitimacy in a Pluralistic, Liberal Society, 111 ETHICS 344 (2001).
(161) This problem exists even if the crime is complete; acts often do not speak for themselves. See WILLIAMS, supra note 153. Nonetheless, the further down the line of Dressler's six steps a person has gone, the more likely it is that her actions will at least be strongly corroborative of one intent over another.
(162) See JOHN RAWLS, The Idea of Public Reason, in POLITICAL LIBERALISM 213-20 (1993).
(163) For a similar view on the need to be sparing in the use of criminal law, see DOUGLAS HUSAK, OVERCRIMINALIZATION: THE LIMITS OF THE CRIMINAL LAW 92-103 (2008) (arguing for what I would call a pro tanto right not to be punished).
(164) This view on flouting the law supports what is sometimes known as weak retributivism. See Sanford H. Kadish, Foreword: The Criminal Law and the Luck of the Draw, 84 J. CRIM. L. & CRIMINOLOGY 679, 698 (1994) (distinguishing strong and weak retributivism).
(165) See supra note 160 and accompanying text.
(166) See JOHN STUART MILL, ON LIBERTY 72-73 (David Spitz ed., W. W. Norton & Co. 1975) (1859) (discussing the propriety of responding to vice by forming an unfavorable opinion of those who have it, rather than by intentionally punishing them; the former has as its natural consequence the loss of valuable social status and opportunities on the part of those who are thought of as vicious, but when such a loss is not intentionally created to punish, it can be more easily justified).
(167) Chesney, supra note 138, at 435.
(168) See MINORITY REPORT, supra note 157.
(169) While that proposition may seem trivially true, it has been contested. See Christopher New, Time and Punishment, 52 ANALYSIS 35 (1992); Daniel Statman, The Time to Punish and the Problem of Moral Luck, 14 J. APPLIED PHIL. 129 (1997). For replies, see Saul Smilansky, A Time to Punish, 54 ANALYSIS 50 (1994), and Richard L. Lippke, No Easy Way Out: Dangerous Offenders and Preventive Detention, 27 LAW & PHIL. 383, 388-91 (2008).
(170) Slobogin, supra note 32, at 4, argues that people who cannot be deterred for this reason ought to be preventively detained. But his argument rests on a faulty premise: "that the person is so lacking in the ... willingness to adhere to society's most basic prohibitions ... that punishment is not warranted." Id. at 5. Slobogin confuses the fact that punishment would have no deterrent effect with the argument that it is unwarranted; it can be warranted for other reasons, such as retributive and expressive ones.
(171) Colman McCarthy, Better Protection for Women in Danger, WASH. POST, June 13, 1992, at A21.
(172) This is the main concern of ALEXANDER & FERZAN, supra note 138, at 212. It is also the concern of all who favor the idea of a locus penitentiae.
(173) See supra note 161 and accompanying text.
(174) United States v. Parr, 545 F.3d 491, 495 (7th Cir. 2008).
(179) See supra notes 135-137 and accompanying text.
(180) MODEL PENAL CODE [section] 2.02(6) (1985); see also WAYNE R. LAFAVE, CRIMINAL LAW [section] 3.5d, at 236-37 (3d ed. 2000).
(181) What if the person is in jail and forms the intention to commit a terrorist act if he ever gets out? There may not be much he can do, beyond dreaming up more or less detailed plans to act on the intention. In that case, it may turn on whether he can expect to be released or not. If, like Parr, he expects his imminent release, then his conditional intention should count as if it were an unconditional one. If he expects never to be released, then his intention is basically a fantasy. The harder question is what to do with an intention that the person expects to have a chance to act on, but not for many years. The fight answer may then depend on other factors that show whether the intention is sincerely adopted or allow the possibility that it is a mere fantasy.
(182) ALEXANDER & FERZAN, supra note 138, at 201-10.
(183) Id. at 200-0l.
(184) Id. at 202-03.
(185) See supra notes 149-150 and accompanying text.
(186) See DRESSLER, supra note 100, at 431 ("It is said that an agreement to commit a criminal act is concrete and unambiguous evidence of ... the firmness of [the actor's] criminal intentions....").
(187) See, e.g., United States v. Parr, 545 F.3d 491,498 (7th Cir. 2008) ("To assess whether Parr's statements were true threats, the jury needed to make inferences from the background and context about his demeanor at the time he made the statements--to decide, under the circumstances, whether he conveyed the impression that he was serious or joking.").
(188) See supra notes 180-181 and accompanying text.
(189) ALEXANDER & FERZAN, supra note 138, at 203-06.
(190) Id. at 204.
(191) Id. at 205. The authors also complicate their example by supposing that the actor might blow up the stadium when it contains only "a dangerous terrorist cell." They seek thereby to introduce the complication that her act might not be culpable at all.
(192) Id. at 206-08.
(193) See supra notes 140-145 and accompanying text.
(194) ALEXANDER & FERZAN, supra note 138, at 207.
(196) Renunciation, as described by the Model Penal Code, is a problematic affirmative defense. It allows one to negate the earlier criminal act of forming and acting on a criminal intention. See MODEL PENAL CODE [section][section] 5.01(4) (renunciation of an attempted crime) and 5.03(6) (1985) (renunciation of a conspiracy). As Stephen Morse wrote, renunciation "is an anomalous defense because the criminal who is prima facie guilty of inchoate crime has satisfied the elements of the crime and acted without justification or excuse." Morse, supra note 26, at 288. Morse thinks the right interpretation of this anomalous defense is that the crime hasn't "really" been committed until abandonment is no longer possible. Id at 289. As I explain immediately infra, I think the better view is either that it is simply an anomalous defense, or, if that anomaly is unacceptable, that it should not be taken to be a complete defense.
(197) See DRESSLER, supra note 100, at 411 (pointing out that "many courts today continue to decline to recognize the defense"). He goes on to say that only objectivists reject the defense. Id. at 412. The position I take in the text offers a subjectivist reason for rejecting the defense, at least as a complete defense.
(198) ALEXANDER & FERZAN, supra note 138, at 208-10. Morse also makes this point: "The affirmative defense of abandonment or renunciation, which is applicable only to inchoate crime, is further evidence that criminalizing such conduct is a preemptive response to potential danger rather than a retributive response to past wrongdoing." Morse, supra note 26, at 288.
(199) ALEXANDER & FERZAN, supra note 138, at 208.
(200) Again, see supra notes 140-145 and accompanying text.
(201) As Dressier points out, the defense of abandonment applies "only if the defendant voluntarily and completely renounces her criminal purpose." DRESSLER, supra note 100, at 411.
(202) Alexander and Ferzan accept attempt liability only for "complete-but-imperfect" crimes in which "the actor performs all of the acts that she set out to do, but fails to attain her criminal goal." Id. at 379.
(203) Indeed, the reverse leads to a perverse conclusion, which I have noted before. Walen, supra note 145, at 375 n.6 (arguing that the failure to recognize the culpability of an attempted crime that does not reach the last required act means that someone who chases someone else through a crowd trying to shoot him but never getting a clear shot is no more guilty of a crime than someone who runs through a crowd with a toy gun trying to scare others as a stupid practical joke).
(204) See supra note 25 (discussing the case of Abdallah al-Ajmi).
ALEC WALEN *
* Alec Walen, Associate Professor of Law, Rutgers School of Law, Camden; Associate Professor of Philosophy, Rutgers University, New Brunswick. I am indebted to Julio Navarro for excellent research assistance.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Symposium: Preventive Detention|
|Publication:||Journal of Criminal Law and Criminology|
|Date:||Jun 22, 2011|
|Previous Article:||Deadly dilemmas III: some kind words for preventive detention.|
|Next Article:||The status quo bias and counterterrorism detention.|