Printer Friendly

Breaking the law to enforce it: undercover police participation in crime.


 A. The Difference Between Undercover and Conventional Policing
 B. The Contemporary Significance of Undercover Policing
 C. Types of Undercover Policing
 1. Surveillance
 2. Prevention
 3. Facilitation
 D. Participation in Crime
 1. Providing opportunities
 2. Maintaining cover and access
 3. Rogue cops

 A. Direct Liability
 1. Mental state requirements
 2. The public authority defense
 B. The Prosecution of Targets
 1. Defenses raised by targets
 a. The entrapment defense
 b. Due process limits
 2. Eliminating barriers to conviction
 C. Internal Guidelines
 D. Authorized Criminality in Comparative Perspective

 A. Transparency and Police Action
 B. Unfettered Discretion
 1. Police discretion and democratic policing
 2. Discretion and authorized criminality
 C. Moral Ambiguity
 1. Moral uncertainty and the undercover officer
 a. The stress of deception
 b. The harm of engaging in authorized crime
 2. Moral authority and the community

 A. Increasing Transparency for Undercover Operations
 B. Limiting Policing Discretion in Undercover Work
 C. Expanding the Research Agenda Beyond Criminal Procedure



Covert policing necessarily involves deception, which in turn often leads to participation in activity that appears to be criminal. In undercover operations, the police have introduced drugs into prison, (1) undertaken assignments from Latin American drug cartels to launder money, (2) established fencing businesses that paid cash for stolen goods and for "referrals," (3) printed counterfeit bills, (4) and committed perjury, (5) to cite a few examples. (6)

In each of these instances, undercover police engaged in seemingly illegal activity to gather evidence or to maintain their fictitious identities. Yet unless these acts are committed by "rogue cops" not authorized to participate in illegal activity, these activities aren't considered crimes. Indeed, they are considered a justifiable and sometimes necessary aspect of undercover policing.

This practice of authorized criminality is secret, unaccountable, and in conflict with some of the basic premises of democratic policing. (7) And to the extent that authorized criminality presents mixed messages about their moral standing, it undermines social support for the police. (8) While the practice isn't new, authorized criminality raises fundamental questions about the limits of acceptable police conduct and has been too long ignored.

What is authorized criminality? I define it as the practice of permitting covert police officers (9) to engage in conduct that would be criminal (10) outside of the context of the investigation. (11) We can then distinguish it from other covert policing tactics, such as passively deceptive surveillance, or the police adoption of the role of a victim rather than that of a fellow criminal. (12) Excluded too are instances where police may cross ethical boundaries but not legal ones, such as when undercover investigators stage homicides or other fictitious violent crimes in hopes of building credibility. (13) In Part I, I further situate authorized criminality in the context of covert policing, by discussing how undercover operations differ and why undercover police participate in crimes. While empirical data is limited, the available evidence shows that authorized criminality is a widely used aspect of undercover work.

In Part II, I argue that, despite its widespread use in covert operations, authorized criminality is the subject of little regulation or guidance. In the vast majority of situations, the police are immune from prosecution, so long as their actions lie within the scope of their official undercover role. A legal justification called the "public authority defense" shields these activities from criminal liability. (And the defense is rarely needed because police are very seldom prosecuted. (14)) Other potential sources of regulation, including the entrapment and due process defenses that can be raised by defendants targeted in covert operations, are equally unlikely to regulate authorized criminality in day-to-day practice.

The absence of any meaningful regulation is remarkable because, as I argue in Part III, authorized criminality implicates some of the most fundamental questions regarding the role of police in a democratic society, questions that have captivated legal scholars of the police for the past fifty years. (15) Transparency and rulemaking counterbalance the pervasive and necessary use of police discretion. Yet secrecy and untrammeled discretion characterize the participation of covert police in criminal activity. This has important practical and normative consequences. We do not know much of what covert police do or how they decide to do it. What is known--that the police may in some circumstances act "above the law"--puts the police in a position of moral ambiguity. Enforcement tactics trump concerns about the moral standing of the state. (16)

With these harms in mind, I offer three proposals in Part IV. First, regular public reporting on the frequency and nature of authorized criminality would increase accountability. We know too little about how often and in what circumstances covert police are permitted to participate in crime. The absence of publicly available information about authorized criminality is especially troubling in light of its lax regulation, its consequences, and its increasing importance in terrorism investigations. (17) Greater transparency increases accountability, and can provide us with a basis for determining when participation in crime is not worth its benefits.

Second, the use of administrative guidelines within police departments would curb unnecessarily free discretion when police engage in authorized criminality. Courts and legislatures have expressed little interest in regulating undercover work. And though a few legal doctrines exist to limit police behavior, undercover investigators are too rarely the subject of criminal prosecution for these doctrines to provide meaningful restraint. Courts overwhelmingly deny legal challenges to undercover tactics, albeit with a discomfort exemplified in comments of one federal appellate court: "Undercover police work in general ... is an unattractive business, but that is the nature of the beast...." (18) But this ill-defined notion of necessity usually doesn't involve any consideration of competing concerns, and thus doesn't help regulate authorized criminality in any way. Administrative guidelines, by contrast, can both guide and restrain the police when difficult judgments must be made in the field.

Finally, the scholarly agenda regarding the regulation of the police must venture beyond the confines of the United States Supreme Court's concerns. Legal commentary focuses primarily on constitutional criminal procedure. (19) While undercover policing has long engaged the attention of sociologists, psychologists, and screenwriters, (20) it has failed to capture the sustained interest of legal scholars to the same degree other police practices have. (21) Undercover policing is a marginal legal academic interest. (22) Yet covert policing is rife with "complexity and paradox"; (23) so too is the particular practice of "state sanctioned lawlessness" (24) that takes the form of undercover participation in crime. One explanation for this neglect may be the "pull" of criminal procedure. To the extent that the United States Supreme Court has addressed undercover policing in the investigative stage, (25) it has found the practice to lie outside of the Fourth Amendment's protections. (26) This focus has meant that those police practices left mostly untouched by federal constitutional law lie beyond the focus of the legal academy as well. It isn't obvious, however, that authorized criminality is any less challenging to notions of democratic policing than is racial profiling or excessive force, to take two examples of extensive critical and popular interest.


What is the role of authorized criminality in undercover work, and how is the latter to be distinguished from ordinary street policing? Scholars have identified several analytically useful ways of categorizing undercover work. This Part uses these categories to provide an introduction to undercover policing and the place of authorized criminality in it.

A. The Difference Between Undercover and Conventional Policing

At first glance, it may seem that the key distinction between undercover work and all other kinds of policing is deception. Deception is used, however, in many aspects of policing. (27) The detective may lie to the defendant in order to gain a confession. A uniformed officer might con an armed and barricaded suspect into providing entry by promising no arrest. And so the fact is that petty deceptions pervade the craft of effective policing. (28) The difference between these deceptions and those of undercover work may be a matter of a degree, but it is a significant one. A detective may lie in the interrogation room about the status of a case to encourage a confession: a deception of purpose. In undercover work, suspects are unaware of both the purpose and the identity of the police. (29) Indeed, the objective of undercover policing is to capture criminals in their "natural" state, although of course the irony is that the observers are duplicitous, or, in the cases of bait-sales and street crime decoys, (30) are part of the circumstances of the crime.

B. The Contemporary Significance of Undercover Policing

Investigative deception is a firmly entrenched aspect of contemporary American policing. Even critics of undercover work generally acknowledge that its elimination is neither feasible nor desirable. (31)

And two related historical developments suggest continued, if not greater, reliance upon undercover policing. First, the increasing complexity of the United States Supreme Court's criminal procedure cases in the post-Warren Court era (32) exerts a "hydraulic pressure" on the police to use techniques that the Court has chosen not to regulate as heavily as it has with regard to searches and seizures of homes, cars, and people. (33) Nowhere is this more explicit than in the Court's creation of the third-party doctrine. In a series of cases, the Court has emphatically denied Fourth Amendment protection to those who, while under police investigation, have disclosed information to third parties, whether that third party is a true criminal associate, a police informant, or an undercover investigator. (34) Criminals assume a risk that their friends are not allies at all, and the police retain a powerful investigative technique where no warrant or any other prior justification is necessary. In addition, the Court has held the Miranda warnings inapplicable in the context of undercover interrogations. (35)

Second, over the past fifty years, the police have gradually deemphasized physically coercive techniques in favor of others that emphasize psychological coercion or deception. Evidence once obtained by the "third degree" or other similarly brutal tactics is neither tolerated by legal constraints nor social mores. (36) Deception, whether in an investigation or the interrogation room, is one of the tools the police have come to rely upon in the place of brute force.

C. Types of Undercover Policing

Unlike an impulsive or opportunistic crime, some crimes involve secretive, complex, and consensual activities. The manufacture of methamphetamine, (37) the bribery of local officials, (38) food stamp fraud, (39) prostitution, (40) dogfighting rings, (41) and, at one time, homosexuality, (42) are examples of such offenses, and they are difficult, if not impossible, to investigate if the police must wait for victim complaints, witness statements, or physical evidence. (43) If these crimes are to be prosecuted successfully, then the police must infiltrate criminal ranks or play willing victims.

Such undercover operations are not the specialty of a few departments, but are instead used widely among police departments of varied sizes. Likewise, undercover operations are usually used as an initial course of action rather than as a means used when others have failed. (44) A leading scholar of undercover policing, sociologist Gary Marx, identifies three different types of undercover investigations, distinguished by their varying objectives: (1) surveillance or intelligence operations, which are the most passive activities, followed by (2) preventive operations, which take a more active approach, and (3) facilitative operations, which require the most active involvement of the police. (45)

1. Surveillance

Surveillance operations use deceptive techniques to gather information about completed, ongoing, or planned crimes. The undercover agent's primary role is to gather information, rather than to influence events. Most surveillance operations are anticipatory rather than postliminary. (46) Thus, while some undercover investigations seek missing persons or goods (i.e., crimes that have already taken place), most target crimes that have not yet occurred. Undercover agents may be sent into various settings--prisons, schools, bars, or other institutions where malfeasance is suspected--and be instructed to look out for suspicious activity. (47) Given the limited ambition of the operation, the possibilities for impermissible police encouragement--entrapment--of targets is less likely in these investigations than they are in facilitative operations. (48)

2. Prevention

Requiring more action than surveillance investigations, preventive undercover activities seek to stop an offense from taking place at all, or at the very least, make its commission much more difficult. Prevention may take the form of weakening or diverting the suspect: an undercover agent planted in a political demonstration advocating violence may try to defuse the crowd by arguing for nonviolence. (49) Alternatively, the operation may focus on strengthening victims ("target hardening"): a law enforcement agency may advertise "get rich quick" schemes to lure unsuspecting customers for the purpose of providing them with warnings and advice against future fraud. (50)

3. Facilitation

In contrast to preventive operations, facilitative ones attempt to encourage the commission of an offense, either through strengthening suspects or by weakening potential victims. This may be done by the provision of aid, encouragement, goods, resources, or markets for the suspect. The role that undercover agents play in facilitative operations depends on whether they are posing as accomplices or as easy victims. In the former case, cops play the willing car thief, fence, hit man, or corrupt politician. (51) In the latter, police may pose individually as decoys for assaults or pickpocketing, or in more complex investigations agents may set up a house of prostitution or a business ripe for extortion. (52)

The nature of facilitative operations has changed over time. While traditional covert investigations involve targeted policing based upon intelligence, covert policing has expanded to include more diffuse and open-ended investigations premised on probabilities and temptations, and thus without specific suspicions, complaints, or suspects. (53) Today, not only are facilitative operations aimed at traditional vice crimes such as narcotics use and prostitution, they have also expanded into areas such as the "integrity testing" of public officials. (54)

Of the three types of covert policing outlined here, facilitative operations are the most controversial. The possibility of police entrapment is most likely in a scenario where the police are actively encouraging crime. And whether or not a particular investigation meets the high hurdle of legal entrapment, (55) the conscious decision on the part of the police to create "opportunity structures" for the commission of crimes leaves many uneasy. Police may pose as motorists to catch extortionist traffic police, as small business owners vulnerable to shakedowns by health inspectors, and as corrupt politicians agreeable to influence. (56) While surveillance and preventive operations have analogues in conventional policing, facilitative operations attempt to maintain a fine balance of creating criminal opportunities in order to impose crime control. (57) From the police perspective, facilitative operations use fewer resources and produce the necessary evidence and arrests more quickly than a surveillance operation can.

Facilitative operations also raise the serious issue of crime amplification: the possibility that the very undercover investigation meant to catch criminals in the act may actually produce more crime. (58) Crime amplification can refer to the crime targeted by the investigation; would it have occurred but for the existence of the investigation itself The concerns of crime amplification can, however, extend much more broadly: what unintended criminal consequences did the facilitation produce? In the latter category, facilitative operations may generate the following effects: the production of black markets that generate funds for more crimes; the introduction of ideas, motives, and confidence for further offenses; and the presentation of attractive temptations to those not specifically targeted by the investigation but who nevertheless take advantage of the opportunity created by the police. (59)

D. Participation in Crime

Undercover officers participate in authorized crimes for a number of different reasons. Two of the most important are: (1) to provide opportunities for the suspects to engage in the target crime, and (2) to maintain a false identity or to facilitate access to the suspect. These needs are at their greatest in facilitative operations, when police must both maintain their covert identities as well as encourage the commission of crime (short of entrapment). As discussed in Part II, the police may legally participate in crime so long as the conduct furthers legitimate objectives. When undercover officers stray from crime control objectives and participate in crime, however, these "rogue cops" leave the bounds of authorized criminality and become mere criminals themselves.

1. Providing opportunities

In facilitative operations, the police furnish a simulated environment (60) that can be as elaborate as the establishment of a false business or as simple as the presentation of a false identity as an especially vulnerable victim. (61) Many of these facilitative activities would constitute crimes had the police agents not been given the authorization to commit them.

In the most commonplace stings, the police may pretend to be drug users or illegal gun buyers looking for a willing seller. (62) In variations of "reverse stings," undercover officers may provide the illegal drugs themselves, the chemicals necessary for drug manufacture, or the "buy" money to the suspects. In United States v. Russell, for example, the United States Supreme Court upheld the government's supply of a chemical (phenyl-2propanone) to the defendant so that he could use it to manufacture methamphetamine. The Court observed that "the infiltration of drug rings and a limited participation in their unlawful ... practices" is a "recognized and permissible means of investigation...." (63)

2. Maintaining cover and access

Police participation in crime can also play an important part in maintaining an officer's covert status. (64) Criminals try to flush out suspected undercover investigators by testing their willingness to engage in crime. (65) A reluctance to participate on the part of an apparent criminal associate can "jinx a deal and arouse suspicion" in a covert investigation. (66) Without the police playing their fictitious roles as closely as possible, criminals could easily exclude those suspected of infiltrating their ranks simply by refusing to tolerate passive behavior. (67) The need for such participation is well recognized. For instance, in affirming the conviction of a defendant over his objections to an undercover officer's participation in drug use, an Ohio appellate court stated that "an undercover agent engaged in the business of trying to stamp out the illicit drug traffic may smoke marijuana in order to give the appearance of validity to his conduct." (68)

This participation in crime may not always dampen suspicions. A suspicious drug dealer may pride himself upon identifying the signs of a "narc": a disheveled appearance that nevertheless looks staged, a physical bearing that betrays the quasi-militaristic culture of the police, and eyes that look "alive" and untouched by real addiction. (69) Authorized criminality is just one of the dramaturgical tools needed by the undercover officer for maintaining the deception. (70)

And participation in crime may not always be triggered by suspicion. The police may decide that participation in offenses increases access to a target, even if an agent's false identity is not in question. Thus, for example, a deep cover operative in a criminal organization may deem it necessary to participate in crimes in order to rise in the hierarchy and gain access to the organization's upper echelons. Or, undercover officers may conduct buy and busts in order to "flip and roll" drug dealers, i.e., turn them into informants. (71)

3. Rogue cops

Finally, the isolation, stress, and psychological toll of undercover work (72) sometimes lead undercover investigators to exceed the bounds of authorized criminality altogether, and participate in offenses as ordinary criminals. It is not entirely uncommon for undercover cops to "go native" and believe in the truth of their own fictive identities. (73)

The practice of authorized criminality may contribute directly to this problem. The conceptual line between authorized and unauthorized criminality is clear: unauthorized crimes further no law enforcement purpose. In the trenches, however, the difference between pretending to use drugs to maintain cover and using drugs to socialize with "friends" in the criminal underworld may be a difficult distinction to draw, particularly for those investigators who are asked to assume "deep cover" roles in which true identities must remain deeply suppressed for long periods of time. (74) In these situations, the costs and visibility of unauthorized criminal participation are low, while opportunities are pervasive. (75)


Although courts and commentators have acknowledged that the practice of authorized criminality is troubling but necessary, the conditions under which undercover police officers may participate in crime have seldom been the subject of regulatory oversight. (76) Instead, what exists is a patchwork of applicable state and federal constitutional law restraints that loosely regulates undercover operations and generally accepts that undercover officers "violate the letter of the law in order to catch criminals." (77) This Part describes the relevant doctrines--the direct criminal liability of police officers engaged in authorized criminality, limitations on the prosecution of targets caught in undercover operations, and administrative guidelines within police departments--and concludes with a brief discussion of the experiences of other countries.

A. Direct Liability

Instances in which undercover police officers have faced prosecution are rare. (78) Nevertheless, it is possible that a police officer who participates in criminal activity during an undercover investigation might be prosecuted. If such a prosecution arises, mental state requirements and the public authority defense are likely to shield the officer from criminal liability.

1. Mental state requirements

In a number of instances, an undercover officer who participates in criminal activity will lack the mental state of an applicable crime, and so risks no criminal liability. (79) The traditional distinction between general and specific intent crimes illustrates the problem. For example, many (though not all) drug possession offenses require a specific intent to sell or distribute. (80) An undercover officer who only pretends to be a drug seller will lack any such specific intent; the mental state of the offense required for criminal liability does not exist.

2. The public authority defense

In other instances, the police participation will appear to meet the substantive definition of a crime. In the unlikely event that an undercover officer were prosecuted for his participation in crime, the public authority defense, recognized in every American jurisdiction, (81) would justify his actions and relieve him of criminal responsibility. Not limited to the undercover context, the affirmative defense, also known as the law enforcement authority defense, justifies otherwise-criminal conduct when that action is taken by a police officer (or a private person under the direction of a police officer) in order to effectuate an arrest, to stop a fleeing criminal, or to prevent a crime. (82) The defense exists so that the threat of criminal prosecution will not hamper police objectives, but it, like other defenses, has important limitations. The police conduct must be authorized, (83) and the means used by the police must be necessary. (84) Some jurisdictions may also impose an additional proportionality limitation. (85) Thus, for example, a police officer cannot resort to physical force if psychological coercion (such as a "command voice") would suffice; (86) nor could a police officer shoot (i.e., use deadly force against) a fleeing pickpocket. (87)

Since the public authority defense permits the police to engage in otherwise illegal conduct for legitimate law enforcement purposes, then it certainly should apply to the undercover context. (88) This "wholesale" immunity is not limited to the commission of particular crimes, nor is it limited to special categories of police personnel. (89)

As with self-defense and other criminal law justifications, the law enforcement defense exists to justify conduct that otherwise meets the elements of a criminal offense because of an overriding principle. Necessity and proportionality may be readily determined in the use of force by the police, but in the undercover context such determinations are more difficult. How, for instance, should we make a necessity determination when the police participate in a money laundering scheme and the reason for such participation is the success of a long-term undercover operation whose exact parameters have not yet been determined? Not every use of covert policing exhibits these ambiguities, but they do exist in long-term, facilitative, and open-ended investigations, where concerns about police use of authorized criminality are greatest. (90)

Whatever its conceptual underpinnings, the limits of the public authority defense have not been rigorously tested. Instances in which undercover police have used this defense are rare because they are seldom, if ever, prosecuted. (91) Prosecutorial discretion thus bolsters the "blanket" immunity of undercover investigators. (92) Most undercover police can be confident that they will not be prosecuted, and may indeed receive explicit assurances from the local prosecutor before the operation even begins. (93) For practical purposes, then, the use of the defense for authorized criminality largely exists as a scholarly curiosity. (94)

B. The Prosecution of Targets

Police engage in authorized criminality to obtain the necessary evidence for the successful prosecution of their targets. The entrapment and due process defenses, however, provide limitations on what the police may do. A successful defense raised by the target will bar prosecution. Courts have not interpreted these defenses very strictly, however. In addition, the trend of modem substantive criminal law has been to encourage undercover activity and discourage claims by defendants that they have committed no offense.

1. Defenses raised by targets

a. The entrapment defense

In cases where the target of the investigation claims that he was illegally induced by the police into committing a crime, the target can raise an entrapment defense to his prosecution. Raising the entrapment defense is a claim that one has, for instance, been bullied by an undercover agent into selling thousands of pseudoephedrine pills, (95) encouraged to solicit sex from an apparently underage girl on the Internet, (96) or cajoled into providing liquor to an undercover officer during Prohibition. (97)

The doctrine of entrapment is really a means to identify those facilitative investigations (98) that have crossed a line between the permissible and impermissible police encouragement of crime. (99) The judicially created doctrine imposes a limitation of reasonableness on the use of undercover techniques so that otherwise innocent persons are not unfairly selected and then pressured into committing a crime. (100) In the so-called "subjective" approach to entrapment that is used in a majority of jurisdictions, the police cannot induce a target who was otherwise not personally disposed to commit the offense. (101)

But the entrapment defense poorly regulates authorized criminality. A successful entrapment defense immunizes a criminal defendant from prosecution, (102) and so puts the police on notice as to what conduct will thwart a successful prosecution in the future. (103) But because the entrapment defense focuses primarily on the defendant's predisposition to criminality rather than the level or degree of police encouragement, the doctrine has not prompted courts to devise a "meaningful definition of what constitute[s] impermissible participation in the offense" (104) by the police. Most instances of police participation will not constitute entrapment (105) so long as the defendant was a ready and willing criminal. (106) The police may also deliberately thwart an anticipated defense. For instance, an undercover officer may act specifically with the intent to undermine an anticipated entrapment defense by, for instance, developing conversations with the target that demonstrate very explicitly the target's personal motivations for committing a crime. (107) In sum, because successful entrapment defenses are relatively rare, (108) they pose an impractical source of regulating police behavior. (109)

b. Due process limits

Even in cases where the defendant has failed to prove entrapment (because of his predisposition to commit the offense), (110) the conduct of the police may nevertheless in some cases violate due process rights. (111) The standard used in the due process analysis varies, (112) but the basic principle underlying the claim in successful cases is that the police conduct has violated basic principles of justice and fairness. (113)

One example: Two undercover officers repeatedly tried but failed to obtain cocaine from their targets, Stanley Robinson and Bobby Shine. Robinson, tired of being badgered at his local bar, sold the officers a bag of sugar. Enraged by the deception, the undercover officers, after drinking for several hours, entered Robinson's home in the middle of the night with a brandished weapon and a demand for drugs or money. Searches of Robinson and Shine at gunpoint yielded two bags of cocaine. In reversing Shine's conviction, the state appeals court noted that this was "one of those rare cases" where police conduct constituted a denial of due process. (114)

But such a case is exceptional. In the undercover context, such claims of outrageous government conduct rarely succeed. (115) Generally, judges are reluctant to impose restrictions on the lengths the police may go in undercover operations by participating in authorized crimes. Some judges have expressed the view that there is, in theory, a limit to what the police can do, but few, if any, cases are found to cross that line. (116) Even when they do, the due process standard doesn't guide future police action particularly well. If the police may not manufacture crack cocaine for use in reverse stings because it is "outrageous," (117) why does the sale of "ordinary" cocaine in a reverse sting pass muster? (118) No clear standard defines acceptable from unacceptable police conduct.

The due process defense is unlikely, then, to regulate authorized criminality in any effective way. Like the entrapment defense, a successful due process defense immunizes the defendant from criminal liability, rather than sanctioning the police officer directly. As for the police whose conduct is considered so outrageous that it violates due process, the due process defense doesn't add more regulatory oversight than that provided by the public authority defense. Presumably, undercover officers engaging in outrageous conduct are not acting in an "authorized" manner, (119) and consequently face criminal liability through direct prosecution. (120)

2. Eliminating barriers to conviction

As the previous sections show, a defendant can rarely raise an entrapment or due process defense with success. Moreover, the trend of American criminal law has been one of removing barriers to conviction in the undercover context. The Model Penal Code's reform approach to inchoate crimes reflects this view. (121)

At common law, the defense of "legal impossibility"--but not "factual impossibility"--permits a defendant charged with attempt to avoid criminal responsibility. (122) Thus a person could not be guilty of attempting to shoot a stuffed deer out of deer season in the belief that it was alive--a case of legal impossibility--but could be guilty of attempted murder by shooting into the empty bed of his victim--a case of factual impossibility. (123) Under the traditional approach, an undercover sting might prevent the conviction of the defendant for attempt, as in the famous case of Booth v. State. There, the defendant successfully raised a legal impossibility defense in his prosecution for an attempt to receive stolen property; the police interception of the coat before it reached Booth's hands had rid the coat of its "stolen" character. (124) The Model Penal Code's elimination of any impossibility defense resolves the difficulty many jurisdictions faced in distinguishing between the two categories. (125) Booth would have been guilty under the Code. (126)

This is also the case with the crime of conspiracy. The bilateral requirement of common law conspiracy means that there must be at least two people who agree to conspire, thus barring conviction where the target "conspires" with an undercover officer. (127) As with attempt, the Model Penal Code facilitates convictions in these situations by eliminating the bilateral requirement. (128)

In these two respects, the Code, a reform project of the American Law Institute, (129) embodies an approach seeking to make apprehension and prosecution of persons through undercover operations easier. And the approach is consistent with the Code's general aim to punish those who have demonstrated an antisocial mental state. (130)

These developments in substantive criminal law, along with the relatively infrequent use of criminal prosecutions for police, reflect a permissive attitude and "largely instrumental focus" (131) in American law regarding undercover policing and authorized criminality in particular. Neither legislatures nor courts have shown any enthusiasm for regulating authorized criminality in any significant way.

C. Internal Guidelines

Internal departmental or agency guidelines provide another source of potential control over authorized criminality in undercover operations. (132) At the federal level, the Department of Justice refers to the Attorney General's Guidelines on Federal Bureau of Investigation Undercover Operations, (133) most recently revised in 2002. (134) "Ordinary" undercover operations require prior approval by the Special Agent in Charge of each FBI office, based on a written determination that the proposed investigation will be effective and that it will be conducted in a minimally intrusive way. (135) When especially sensitive circumstances exist, such as the proposed investigation of public officials, media organizations, or an alleged terrorist organization, an Undercover Review Committee consisting of Department of Justice and FBI officials must approve the proposed undercover operation. (136) Ordinarily, an authorized undercover operation may last up to six months, subject to a six-month renewal, and involve expenditures of no more than $100,000. (137)

The Guidelines also explicitly consider the involvement of FBI agents in illegal activity during the course of an undercover operation, referred to as "otherwise illegal activity": "any activity that would constitute a violation of Federal, state, or local law if engaged in by a private person acting without authorization." (138) In an ordinary undercover operation, the Guidelines explicitly forbid undercover officials from participating in violent acts except as a matter of self defense, encouraging criminal activity in a manner that would constitute legal entrapment, (139) and using illegal investigative techniques such as illegal wiretapping. (140)

The Special Agent in Charge may, however, authorize undercover FBI agents to participate in certain offenses such as the payment of bribes, the purchase of stolen or contraband goods, money laundering (though not more than five transactions not to exceed $1 million), the controlled delivery of drugs (so long as they do not enter commerce), and the making of false representations to third parties. (141) Any official providing authorization for illegal activity must consider that illegal activity is justified only if it is needed to obtain evidence that is not otherwise "reasonably available," to establish or maintain a secret identity, or to prevent death or serious bodily injury. (142) Felonies not specified in the Guidelines are considered sensitive circumstances that must be approved by the Undercover Review Committee as well as the FBI Director, Assistant Director, Deputy Director, or Executive Assistant Director, depending on the circumstances. (143) In all cases, the Guidelines mandate that all "reasonable steps" be taken to minimize the participation by FBI agents in illegal activity. (144)

There is, however, one important limitation to the Guidelines: they are non-binding. (145) The Guidelines, meant for "internal DOJ guidance" only, state that they
 are not intended to, do not, and may not be relied upon to create
 any rights, substantive or procedural, enforceable by law by any
 party in any matter, civil or criminal, nor do they place any
 limitations on otherwise lawful investigative or litigative
 prerogatives of the Department of Justice. (146)

Thus while a violation of the Guidelines may reflect a breach of agency policy, the Guidelines lack the regulatory teeth of a statute or judicially created doctrine that might impose sanctions on the police or provide a defense to a target of an undercover investigation.

Internal guidelines at the federal level, moreover, do not provide a representative picture of administrative rulemaking over covert policing. American policing takes place primarily at the local, and not the national level. According to recent figures, local and state police agencies employ about 770,000 police officers, (147) compared to, for instance, only 12,242 FBI officers. (148)

At the state and local level, the use of guidelines for undercover operations varies greatly; no systematic collection or review of such guidelines exists. A 1994 survey of eighty-nine police departments around the country yielded varied results. (149) While all the departments surveyed conducted undercover operations, only a subset--sixty-two of them--had written guidelines. Of those with guidelines for undercover work, more than half emphasized procedural rules (such as steps for checking out equipment and the proper method of filling out forms (150)) rather than authorization issues (such as when or why operations should be initiated (151)). In their conclusion, the study authors expressed "surprise[] to learn that 23 large municipal police agencies using undercover police work did not have written guidelines" at all. (152)

D. Authorized Criminality in Comparative Perspective

American law is aimed at responding to abuses in undercover policing, but not at regulating ordinary practices. (153) But not all democratic societies provide such "wholesale" immunity for undercover policing. (154) Consider two sources of comparison:

Italian law begins with the presumption that the practice of authorized criminality is illegal. (155) Rather than provide a wholesale form of immunity for authorized criminality in covert investigations, Italian law provides for statutory exemptions that shield Italian police from criminal liability on a crime-by-crime basis. (156) Unlike their American counterparts, then, Italian undercover investigators face a real risk of prosecution if their participation in crime is not expressly authorized by statute, even if the police conduct is meant to further a law enforcement objective. (157) For instance, a specific statute insulates undercover police from criminal liability for participating in drug buys. (158) A reluctance to provide undercover police broader immunity for illegal acts reflects a concern that to do so would "corrode the rule of law." (159)

In Canada, the Parliament addressed authorized criminality directly in response to a 1999 decision of its supreme court, R. v. Campbell. (160) The disputed police activity in Campbell was a garden-variety reverse sting: federal police officers offered to sell a large quantity of hashish to the defendants. (161) The Canadian Supreme Court held that absent direct Parliamentary exemption, police were not clearly immune from criminal prosecution for actions that would be otherwise illegal. (162) While the Canadian Narcotic Control Act explicitly permitted officers to possess illegal drugs in covert operations, no such provision existed for their sale. As a direct result of the Campbell decision, some covert operations were suspended or closed to avoid the potential prosecution of officers in covert operations. (163)

In 2001, the Canadian Parliament enacted a law enforcement justification defense to clarify the status of authorized criminality and to set limits upon its use. 164 Section 25 of the Canadian Criminal Code provides for a justification for otherwise illegal acts committed by the police in the course of an investigation. A police officer may commit an action that is otherwise considered illegal so long as it is "reasonable and proportional in the circumstances." This assessment should take into account factors such as the nature of the authorized criminality, the nature of the operation, and the "reasonable availability of other means for carrying out the public officer's law enforcement duties." (165) While the defense is not limited to specified offenses, excluded from its application are those actions taken by the police which result in "death or bodily harm," constitute a violation of the "sexual integrity of an individual," or a "willful attempt" to obstruct justice. (166) In addition, the statute requires police departments to publish annual reports providing details on the frequency and nature of the acts covered by the defense. (167)


While police, prosecutors, and judges may view authorized criminality in undercover policing as a sometimes unpleasant but practical necessity, the tactic exacts its costs. Some commentators have pointed out that undercover policing by its very nature risks significant social harms because it necessarily involves deception and secrecy. (168) The practice of authorized criminality raises even more serious problems because it adds the perception of criminality to an investigative technique that is already secretive and deceptive. We can't consider the real merits of authorized criminality without weighing the crime control benefits against the harms it visits upon the police and the public's perception of them.

This Part considers three different kinds of harms posed by authorized criminality: the lack of transparency about basic information regarding authorized criminality, the exercise of unfettered police discretion, and the moral ambiguity that arises when police engage in criminal activity in order to pursue criminals.

A. Transparency and Police Action

Police decisions about authorized criminality in undercover operations lack basic accountability because of their largely secret nature. There are no widespread norms regarding systematic data collection on undercover policing generally or with regard to the nature and extent of authorized criminality specifically. (169)

Indeed, those seeking detailed official information on authorized criminality are likely to encounter difficulty. The exemption of federal law enforcement procedures and techniques from the Freedom of Information Act (FOIA) requirements is illustrative. (170) The FOIA provides a person with a legally enforceable fight to obtain access to federal agency records, provided that the information does not fall within one of nine exemptions. (171) Among the Act's nine categorical exemptions is Exemption 7(E), which permits nondisclosure of law enforcement information where it "would disclose techniques and procedures for law enforcement investigations or prosecutions, or would disclose guidelines for law enforcement investigations or prosecutions if such disclosure could reasonably be expected to risk circumventions of the law." (172) The exemption consists of two separate clauses, both of which have been construed broadly by courts to permit non-disclosure of law enforcement procedures, policies, and records. (173) While the second clause ("circumventions of the law") permits withholding of information based on an assessment of harm to law enforcement interests, the first clause is even more expansive, permitting nondisclosure of a law enforcement procedure without any demonstration of a harmed interest.

While the "techniques and procedures" that fall under the scope of Exemption 7(E) have been interpreted as those not generally available to the public, even commonly known techniques have been protected as exempt from FOIA requests when "the circumstances of their usefulness.., may not be widely known." (174) Thus, while undercover policing is a widely known investigative technique, courts have upheld the non-disclosure of information pertaining to specific undercover practices, on the grounds that disclosure of particular techniques would reduce their effectiveness. (175) In fact, nondisclosure has been justified as especially warranted when the technique itself--whether the use of polygraph examinations, (176) bait money, (177) or undercover policing--is meant to operate with some secrecy. (178)

The absence of systematic and specific information is not limited to the undercover context. Although publicly available data is used to monitor many types of executive decision making in modem government, such transparency is often absent in the exercise of police authority. (179) Erik Luna has forcefully argued that a democratic conception of police discretion necessitates the "systemic visibility" of official police actions and justifications. (180) Hiding police decisions from public view, whether or not those decisions are legal or publicly supported, is never benign. (181)

The simple absence of transparency in police decisionmaking can be destructive, both in its potential to breed police abuse as well as to foment public distrust. There is little available public knowledge about the frequency, nature, and conditions of authorized criminality in undercover work. Yet the practice suggests a normative paradox: here the state permits the police to act seemingly "above the law" as they enforce the law. Rather than allay concerns about authorized criminality by holding it up to public scrutiny, the reality is that it is itself a practice under cover. Secrecy suggests that there may be illegitimate reasons to hide the decision making process in this area. Moreover, the potential for abuse is greater when little or no public oversight is available to weigh in upon police decision making.

Finally, cordoning off police decisions from public scrutiny encourages public distrust of the police. As a number of studies of public attitudes toward policing have shown, trust is much more effective as a foundation for public compliance with the law than the threat of punishment or reliance upon personal morality. (182) Public distrust not only conflicts with democratic norms, but a public wary of the police is much less likely to be a legally compliant or cooperative one. (183)

B. Unfettered Discretion

In many instances, undercover officers (and their departments) who participate in authorized crimes also lack significant constraints on their discretion, a situation that lies in tension with basic principles of democratic law enforcement as well as the historical concern about placing restraints on police discretion.

1. Police discretion and democratic policing

The "discovery" of police discretion in the 1950s by police researchers introduced a topic of study and a set of issues that continue to occupy scholars and judges to this day. (184) By itself the concept is uncontroversial; discretion exists whenever two or more choices are available to the decision maker. (185) The police exercise discretion out of necessity. While they may be entitled to exercise their legal authority in many situations, factors that are both practical and symbolic influence what they decide to do in practice. Budgetary concerns can serve as practical constraints, and the attitudes of the community policed can also influence priorities of enforcement. (186) Finally, the culture of the police themselves--a distinct world view emphasizing danger and authority--heavily influences the kinds of persons who are targeted for police interest. (187)

Yet discretion is especially problematic for the police in a democratic society. (188) We ask the police to assume the primary role in enforcing the law and imposing public order, while armed--quite literally--with the ability to rely upon the state's monopoly over legitimate force. (189) At the same time, we expect the police, within the framework of their legal, practical, and symbolic constraints, to exercise their authority fairly. (190)

A great deal of scholarly and practical attention on the police has focused on restricting and guiding this police discretion. Indeed, "police discretion" simpliciter is something of a dirty word, evoking less the motorist let off with a warning than the minority-race motorist stopped for a legally adequate but ethically suspect reason.

Although many of the criminal procedure cases reaching the United States Supreme Court have raised issues of police discretion, the Court has been reluctant to restrict police in their ability to make a variety of decisions in the investigative process. Thus, for instance, the police may: conduct an inventory search of a defendant's car so long as the search is exercised according to some minimal criteria; (191) stop a motorist for a legally adequate reason even if it is not the actual reason for the detention; (192) and arrest someone for a very minor crime if the applicable law permits arrest, even if most police officers would issue only a citation. (193) One area where the Court has reined in discretion is in vagrancy laws so vaguely written that they provide the police with broad license to detain anyone they wish. In its most recent decision of this kind, the Court struck down a Chicago anti-gang ordinance that permitted the police to arrest those on the streets without any apparent purpose. (194)

Ironically enough, the Court's striking down of the Chicago law prompted a robust advocacy of decreased judicial control over police discretion regarding "quality of life" offenses. This policing approach, focusing on the enforcement of minor crimes such as littering and open container laws, (195) has been used on the ground with success in New York and elsewhere. (196) The scholarly debate does not so much focus on the desirability of the "broken windows" approach as question whether a traditional reliance on constitutional void-for-vagueness concerns or a community oversight model should serve as the primary source of constraint over police discretion. (197)

The dominant theme regarding discretion is apparent: we begin with the assumption that wholly unfettered police discretion is undesirable and probably harmful, not just to the individuals who are on the wrong end of unwise exercises of discretion, but also to the public whose support for the police in a democratic society is essential.

2. Discretion and authorized criminality

When should undercover police participate in crimes? Which crimes should they participate in? How many times and for how long should this participation last? Few legal restrictions constrain undercover police regarding the scope of their permissible conduct in the case of authorized criminality as a practical matter. (198) These critical questions are left to individual agencies and departments to decide. (199) The police have considerable latitude over undercover operations, which can range from a straightforward "buy and bust" to a deep undercover operation that may last years (200) and require significant psychological and social adjustments for the officers involved. (201) The applicable legal doctrines--the defenses of public authority, entrapment, and denial of due process--are invoked so infrequently, let alone successfully, in cases of authorized criminality, that as limits they are more theoretical than practical. (202)

Instead, courts often justify authorized criminality by a vaguely defined principle of necessity. For example, if necessity requires a balancing of costs and benefits, few courts consider the potential harms when undercover investigators participate in crime. Typically, courts take the view that "criminal proceedings are not designed to establish the relative equities among police and defendants." (203) While a few opinions have expressed ambivalence about the "unattractive business" (204) of investigative deception while affirming a target's conviction, courts tend not to delve too deeply into the issues raised by undercover policing that have been discussed here. Instead, they frequently find it sufficient to declare that authorized criminality is a necessary though unpleasant evil.

In United States v. Murphy, for instance, the Seventh Circuit considered a prosecution obtained as a result of "Operation Greylord," a complex undercover operation aimed at targeting fixed cases in Cook County, Illinois. The FBI contrived an elaborate undercover operation in which agents staged fictitious cases in the Cook County Courts by posing as defendants and lawyers. In upholding the conviction of John Murphy, a former state judge, the Seventh Circuit rejected his contention that the government's fake cases prohibited them from pursuing crooked judges:
 Murphy's complaint is a more traditional objection to creative acts
 by prosecutors.... In Operation Greylord agents of the FBI took the
 stand in the Circuit Court of Cook County and lied about their
 made-up cases. Perjury is a crime, and Murphy tells us that those
 who commit crimes themselves cannot prosecute others' crimes....
 Bribery ... is a secret act. Because the crime leaves no
 complaining witness, active participation by the agents may be
 necessary to establish an effective case. (205)

Ultimately, the Court's refusal to disapprove of the police tactics in Operation Greylord rests on the view that even if the police tactics appear unpalatable, social disgust is not the appropriate measure: "In the pursuit of crime the Government is not confined to behavior suitable for the drawing room." (206) Other courts have made similar observations. (207)

Likewise, police and prosecutors have deemed it an essential tool for investigation and in some cases superior to the available alternatives, such as exclusive reliance upon confidential informants. (208) This has been a view held even in times when intense public scrutiny has been directed at undercover tactics. In the late 1970s, the FBI initiated an undercover investigation, later known as ABSCAM, in which covert agents and con-man-turned-informant, Melvin Weinberg, posed as representatives of a fictitious sheik seeking favors from public officials--including members of Congress--in exchange for money. (209) Although the investigation eventually resulted in a number of convictions, the FBI's tactics drew controversy and eventually the attention of Congress, which held a series of hearings to examine FBI undercover techniques. (210) In its final report issued in 1982, the Senate Select Committee to Study Undercover Activities, while concerned about the "serious risks to citizens' property, privacy, and civil liberties," as well as to "law enforcement itself' posed by undercover investigations, nevertheless stated that "some use of the undercover technique is indispensable to the achievement of effective law enforcement." (211)

The problem with these justifications, however, is that they extend too broadly. Shielded by an expansive view of necessity, undercover policing enjoys little in the way of searching judicial or legislative scrutiny. Limiting discretion in covert policing is not a priority.

C. Moral Ambiguity

Moral ambiguity is a third significant consequence imposed by the participation of undercover officers in authorized crimes. Those harmed by this moral ambiguity include not only the participating police but also the larger community.

1. Moral uncertainty and the undercover officer

a. The stress of deception

Undercover work exacts many personal costs from individual investigators. Occupational hazards are legion. Not only must the undercover officer present and maintain a credible false identity in a criminal milieu, often he must also gain the confidence of his criminal associates. Accidental disclosure can result in violence against the agent, or, at the very least, the abrupt end of the investigation. These tasks and decisions take place in isolation from other police officers, with the more difficult "deep cover" assignments assumed with even less oversight and less frequent regular supervisory contact. (212) In addition to these risk factors, some departments prefer to use young recruits or relatively inexperienced officers as undercover agents, in part because they pose a smaller risk of recognition by targets. (213) A young detective plucked directly from the police academy and assigned to a two-year deep cover role to investigate Islamic extremists in Brooklyn received no undercover training prior to assignment and remained in contact only with his supervisor, at first only by email. (214)

And the more time spent as an undercover agent, the greater the risk that personal problems will appear. A number of studies have documented the harms visited upon undercover officers: corruption, disciplinary problems, alcohol and drug abuse, (215) interpersonal problems, (216) a "loss of self," and paranoia. (217) In extreme cases, the agents "go native" and become indistinguishable from their targets. (218)

Some of the unique risks and harms to the individual officer can be attributed to the paradoxical nature of the job itself. (219) Because undercover work requires the presentation of a false identity, deception and secrecy are essential skills to the job. Important too is the ability to adapt to the changing demands of the criminal underworld. Thus, successful undercover investigators tend to be those who are especially adept at dissimulation and risk-taking. (220) One undercover officer told a researcher that in deep-cover work in particular the agent "'must have the ability to improvise' because 'there are no rules' to deep-cover work and the person must be 'basically deceitful."' (221) These same officers, however, are also expected to maintain high standards of professional integrity and to avoid temptations to delve into unauthorized activity in the midst of a social setting where the normal constraints of social convention are loosened or wholly absent. (222)

And the irony of this pretense is that it may become reality. Undercover officers can feel torn between actual camaraderie that develops between them and their targets, and the larger purpose for which they have been assigned. (223) Similarly, undercover agents can develop romantic relationships with targets that muddy their priorities. (224) Along with these changed loyalties there may emerge altered attitudes and beliefs, including changed beliefs about the propriety of law itself. (225)

b. The harm of engaging in authorized crime

Permitting agents to participate in crimes adds yet another layer of strain to this tangle of conflicting demands and loyalties by heightening role confusion. Maintaining a dual identity is by itself a difficult task. When the agent is permitted in his official capacity to participate in crime, this may be justifiable and non-criminal as a legal matter, (226) but to the agent, this authorized criminality is, in psychological terms, not a mere simulation. (227) There is camaraderie in a band of thieves; participating in the same crimes as those being investigated increases the risk of over-identification with targets.

Authorized criminality may also contribute to the "moral corrosion" of the undercover agent who is immersed in a world where ethics have already been compromised. (228)

And while the line between authorized and unauthorized crimes may be clear to supervisors, prosecutors, and judges, that is less likely to be the case for undercover agents. To the undercover agent, there may be few sharp distinctions between engaging in crimes to maintain one's cover and those that are simply for self-gain. The temptations to join one's criminal associates are numerous. Working in isolation and secrecy provides opportunities to take shortcuts. The success and thrill of deception can augment a sense of bravado, as well as sow a note of confusion for the agent as to who he "really" is. (229)

While most undercover agents may not deviate from their assigned roles, the dangers are there. These are risks that are inherent to the job of undercover work itself, and authorized criminality exacerbates the problem. Aping the argot, garb, and conventions of crooks already introduces strain to the agent's perspective; to then ask the agent to engage in the criminal acts challenges the agent to ask: who am I?

2. Moral authority and the community

If authorized criminality can unmoor the agent from his moral compass, it may also undermine the moral authority of the police in their relationship with the public. When the police engage not only in investigative deception, but in acts that would otherwise be criminal in the name of crime control, is the "moral tattiness" insufficient to outweigh the benefits of a potentially successful investigation? (230) If, in general, undercover work undermines "the social convention that there is a sharp difference between evil-doers and the righteous," (231) then the particular practice of authorized criminality strains this distinction even further.

Whether or not the action is justified as a matter of legal doctrine, the knowledge that the police are permitted to participate in crime, even for justifiable ends, erodes public trust in the police. (232) To the extent that people react to law in terms of its didactic and expressive functions, authorized criminality transmits a contradictory message. Official police participation in crime, even if to catch criminals, suggests that the state punishes without being what Justice Brandeis called the "potent" and "omnipresent teacher." (233)

And these harms may well outweigh the value of the cases won through undercover means. Public trust and moral authority are essential for the police in a democratic society. Yet the practice of authorized criminality reflects undesirable expressive legal norms. (234) It suggests that criminal wrongdoing is relative or situational, depending on the identity of the perpetrator. It also promotes police behavior that is free of the basic rule-of-law principles that are thought to be basic to democratic policing: public accountability and constrained discretion.


By itself, covert policing raises a host of problems about the optimal mix of effective enforcement tactics and ethical police behavior. The participation in crime by undercover police is a little known and secretive practice that by its very nature challenges core presumptions about democratic policing. When police are permitted to take the additional step of behaving as if they were in fact criminals but for doctrines justifying their conduct, they pose a host of potential harms to themselves, the public trust, and the stability of what it means to enforce the law.

At least three implications follow from this more complete portrait of authorized criminality. First, we should permit much broader public access than is now available to basic information on undercover work, including the use of authorized criminality. Second, one step towards guided discretion would be the use of administrative guidelines. Third, legal scholars of the police must extend their agendas beyond those concerns identified by the Supreme Court, thus drawing attention to neglected subjects like authorized criminality.

A. Increasing Transparency for Undercover Operations

We have greater routine and systematic collection of data on the police today than we did a half-century ago. The federal government collects annual data on arrests, clearance rates, and other police activities from around the country. (235) Yet, as previously discussed, we know little by comparison on undercover activity. (236) There is no systematic collection of data on undercover investigations made publicly available, (237) and in fact much information is deliberately withheld from public view. (238) Instead, researchers must rely upon media coverage, reported cases, the occasional release of official information from government reports and agency press releases, and often sensationalized memoirs from former undercover investigators. (239)

This inattention is unwarranted. Police participation in authorized criminality raises troubling issues of secrecy, unfettered discretion, and moral uncertainty. These are matters no less pressing than racial prejudice, police corruption, excessive force, and other matters that regularly draw public and scholarly attention to the police. The absence of transparency increases the likelihood of public mistrust, hides potential police abuses, and expresses undesirable norms about the moral standing of the police as those entrusted to enforce the law.

The reluctance on the part of the police to disclose such information is understandable. By its very nature, undercover policing tends to draw suspicion. To release details of fictitious identities, police-run brothels and fencing businesses, false intimacies with criminals, and similar sordid details is unlikely to win favor with the public, even if in the legitimate pursuit of criminal activity.

Yet without basic knowledge about the nature and extent of authorized criminality, there is no empirical basis upon which sensible constraints or clear objectives can be crafted. What is the focus of any department's undercover efforts? What is the frequency of and basis for engaging in authorized criminality? Are the police spending the bulk of their time fencing stolen televisions, tendering bribes to public officials, or buying narcotics off the street? One might imagine limiting authorized criminality to certain offenses, or to limiting its use to certain kinds of investigations, but these kinds of street-level decisions should not be made in the abstract. Without knowing basic facts about undercover participation in crime, it is impractical to make substantive decisions about, for instance, which crimes may or may not merit this kind of deceptive practice when balanced against potential harms.

And the need for transparency can accommodate concerns about compromising policing techniques and interfering with pending criminal investigations and prosecutions. Take the example of arrest data once again: we collect a number of details about police arrests, make it publicly available, and do not undermine individual cases or police efficacy as a result. (240) There are good reasons to be more sensitive about undercover work, of course, because it may indeed compromise investigations to divulge very specific information, but these are matters that can be remedied without justifying a wholesale blackout on the collection and publication of less specific information. (241)

Moreover, greater transparency can provide the basis from which regulation can arise. It may be that undercover policing is deemed necessary for the investigation of many crimes, but perhaps not as many as police departments and executive agencies now claim. Even when some crimes may not be as effectively investigated without the ability of police to engage in authorized criminality, a clear understanding of the factual circumstances, including the costs to officers and the kinds of crimes under investigation, may counsel the prohibition of some authorized criminality in light of the potential harms. Transparency fuels informed decision making, which in turn can curb discretion as well as address concerns about public trust.

B. Limiting Policing Discretion in Undercover Work

Undercover police have few restraints on the exercise of their discretion when invited to participate in crime during an investigation. Perhaps the lack of significant curbs on discretion can be attributed to characteristics inherent to undercover policing. Undercover operations can be unpredictable and an investigator may be faced with a sudden invitation to participate in crime. More importantly, though, the legal framework that tells the police what they may and may not do when participating in crime in an undercover capacity isn't particularly useful in the day-to-day operations of the police.

Doctrines like the due process defense and entrapment serve to delineate the outer limits of acceptable undercover police work, but the truly difficult questions have to do with the ordinary situations that police will encounter repeatedly in undercover investigations. We can probably all agree that an undercover police officer cannot participate in murder, but there are a great many other less serious but nevertheless important crimes that the police sometimes do participate in; the due process and entrapment defenses don't particularly help here. Likewise, the public authority defense is no significant curb on discretion; the doctrine doesn't answer the thorniest questions about authorized criminality. Just because undercover officers obtain supervisory authorization to engage in drug sales, forgery, and prostitution does not necessarily mean they have appropriate guidelines in knowing when, how, and to what extent to participate.

The lack of interest shown by the judicial and legislative branches might reflect a concern that neither possesses sufficient expertise to regulate undercover policing. Indeed, George Dix, lamenting the dearth of legal regulation over undercover policing more than thirty years ago, suggested that the "law's failure" to control this investigative technique could be attributed in part to the conscious reluctance of courts and legislatures to delve into matters in which the practical understanding necessary for careful guidance was lacking. (242) But the fear of undue judicial interference has not stopped courts from regulating the police in interrogations, searches, and seizures. In the absence of judicial and legislative intervention, what else might regulate undercover police participation in crime?

Internal departmental guidelines may serve as a starting point. The FBI guidelines are a helpful model; they explicitly acknowledge the practice of authorized criminality and place limits on its use. (243) By requiring explicit authorization in most instances, minimization of police participation in crime, and justification in only limited circumstances, the federal guidelines provide guidance to the FBI prior to any involvement in authorized criminality. Model guidelines might be developed along these lines and provide more guidance than the FBI guidelines do, such as specifying what considerations must be balanced in the extent, degree, and duration of authorized criminality in a given operation.

Not only would guidelines have instrumental value, they would serve symbolic functions as well. Most police officers, when asked about their craft, are unable to justify their actions in terms more precise than "common sense" or "proper action." (244) Regardless of the actual skill involved in police work, such justifications of police work are inadequate in a democratic society. Clear guidelines on police discretion can both dispel some of the morally ambiguous nature of police work as well as assure the public that common problems of police work--which are usually morally and legally complex as well--have been addressed explicitly by the department as a formal matter. (245)

There are good reasons, however, to maintain a healthy skepticism about the likelihood of police-initiated rulemaking in this area. (246) As a practical matter, some have argued that guidelines can be unintentionally helpful to would-be criminals trying to identify police infiltration. (247) Such concerns may justify withholding some information from public view, but not a failure to provide formal guidance over discretion at all.

As a historical matter, police departments have been reluctant to assume the task of internal rulemaking on discretion. While administrative rulemaking was widely embraced in the 1960s and 1970s by reformers and academics alike as a source of control over police discretion, in practice the reaction of police departments has been mixed. (248) Despite the urgings of academics, public policy figures, and some reform-minded police chiefs, rank and file officers often balked at what they perceived to be control by outsiders. (249) For most police departments today, guidelines that do exist tend to focus on internal administrative issues, rather than the problems faced in ordinary police work. (250) When they have been implemented, guidelines have tended to be "crisis-driven" rather than the product of considered reflection. (251) Those institutions in a position to require police rulemaking--the legislatures and the courts--have not used their authority to any significant degree. (252)

C. Expanding the Research Agenda Beyond Criminal Procedure

The lack of substantial data and the reluctant engagement of courts explain a little, but not much, of the curious lack of interest in undercover policing by legal academics who study the police. (253) As a general matter, the police generate a seemingly limitless body of commentary by legal academics. Why has undercover policing and its reliance upon authorized criminality been neglected if, as I have argued, such practices contradict or undermine basic premises of democratic policing?

Part of the answer may lie in the pull of the Court's criminal procedure jurisprudence, and its considerable influence over the research agenda of legal academics. The Supreme Court has played a central role in regulating police investigation through its decisions on the Fourth and Fifth Amendments, which concern searches, seizures, and interrogations of suspects. This regulation has been so considerable that constitutional law, rather than federal or state lawmaking, is the primary source of regulation in these areas. (254)

The central role of the Court in regulating police procedure has had a number of undesirable consequences. As William Stuntz has argued, the aggressive constitutional regulation of procedure has made it more costly for legislatures to regulate policing, and thus they have turned their attention to areas left largely untouched by the Court: substantive criminal law and noncapital sentencing. (255) Yet if criminal justice is to be "representation-reinforcing," that structure is illogical. (256) The Court regulates in areas where legislatures are likely to be most responsive and democratic, and leaves beyond the scope of constitutional law matters such as crime definition and police discretion where concerns about discrimination on the basis of race or wealth are especially high. (257) This distorting effect has led, in Stuntz's view, to skewed legislative attention (and spending) on those areas left largely unregulated by the Court, which in turn has led to over-criminalization and ever harsher sentencing policies. (258)

This too may be the case with the legal scholars of the police and their research agendas. Taking its cue from the Court, legal scholarship has taken up many thorny issues of policing left open, unresolved, or problematic by the Court's Fourth and Fifth Amendment cases. These questions deserve the attention legal scholars of the police have paid them, but that attention has come at the cost of scholarly attention to areas where the Court has paid very little attention: undercover policing, police discretion, and police corruption, to name a few. (259)

Legal scholars of policing experience a pull from constitutional criminal procedure that obscures or overshadows other areas of policing that raise not only doctrinal problems but also fundamental, rule-of-law type concerns as compelling as those addressed in the context of police seizures or interrogations. A roughly comparable dilemma in law and society scholarship is instructive. In a seminal article, Austin Sarat and Susan Silbey warned that an uncritical attempt to address the "pull of the policy audience" tended to distort scholarship. (260) By addressing problems primarily in the "scientific" perspective and terms demanded by the policy audience, law and society scholarship is diminished and critical opportunities are lost. (261) Scholarship exclusively shaped by the policy audience encourages an uncritical acceptance of questions, premises, and objectives of the policymakers. (262)

This problem is a species of agency capture, and criminal procedure scholarship isn't beholden to the Supreme Court in quite the same way. But Sarat and Silbey offer a larger insight about scholarly attention with application here. Taking cues from the Court about the key issues in police regulation tends to reinforce the idea that these are the issues to be addressed as a matter of scholarly interest. Matters left unregulated by the Court are left to the periphery by scholars as well.

But legal scholars of the police should not be constrained by the pull of constitutional criminal procedure. Searches and seizures by beat cops and interrogations by detectives, while important, do not represent the entire spectrum of police behavior. Renewed scholarly attention can raise fresh insights about police problems outside of the dominance of constitutional criminal procedure. Undercover policing, unregulated by constitutional criminal procedure, ignored by legislatures, and marginalized by academics, has been a victim of this neglect.


Investigative techniques can't be measured by their ability to secure convictions alone. Covert operations are an important tool of the police, but the unrestrained use of deceptive practices should make us as concerned as a proposal for total and pervasive surveillance would. The participation of undercover officers in criminal activity should give us pause. Even the appearance that the police are in some instances above the law is troubling. Over time, we have decided that some police tactics cannot be countenanced in a democratic society, whatever their instrumental value. It may not be possible to eliminate authorized criminality, but we should remain alert to its potential for harm.

(1.) United States v. Wiley, 794 F.2d 514, 515 (9th Cir. 1986).

(2.) This kind of police activity was central to Operation Casablanca, a multi-year undercover operation run by the United States Customs Service to target large-scale money laundering of drug trafficking proceeds conducted through Mexican banks. See David Rosenzweig, Laundering Scheme Detailed by U.S., L.A. TIMES, June 2, 1998, at B2.

(3.) Brown v. State, 484 So. 2d 1324, 1325-26 (Fla. Dist. Ct. App. 1986).

(4.) United States v. Gonzales, 539 F.2d 1238, 1239 (9th Cir. 1976) (undercover agents purchased ink, supplies, and a press for a counterfeit operation); see also United States v. Reifsteck, 535 F.2d 1030, 1035 (8th. Cir. 1976).

(5.) See discussion of Operation Greylord, infra Part III.C.2.

(6.) Participation in other illegal activities has been well documented. See, e.g., Anchorage v. Flanagan, 649 P.2d 957, 959 (Alaska Ct. App. 1982) (engaged in sexual acts with prostitutes); Gary T. Marx, Under-the-Covers Undercover Investigations: Some Reflections on the State's Use of Sex and Deception in Law Enforcement, CRIM. JUST. ETHICS, Winter-Spring 1992, at 13, 15 [hereinafter Marx, Under-the-Covers] (established brothels); Jacqueline E. Ross, Impediments to Transnational Cooperation in Undercover Policing: A Comparative Study of the United States and Italy, 52 AM. J. COMP. L. 569, 569-70 (2004) (stole fine art); Alan Dershowitz, Sting II: Police Departments Get into the Act, BOSTON HERALD, June 25, 1982, at 25 (commissioned and financed an "obscene film"); cf. Hampton v. United States, 425 U.S. 484 (1976) (supplied heroin to defendant and participated in its sale to another government agent); Shaw v. Winters, 796 F.2d 1124 (9th Cir. 1986) (sold food stamps and claimed they were stolen); United States v. Parisi, 674 F.2d 126, 127 (1st Cir. 1982) (provided the food stamps that formed the very basis of the conviction for improper use of food stamps); Chaney v. Dep't of Law Enforcement, 393 N.E.2d 75, 79 (Ill. App. Ct. 1979) (forged documents); State v. Putnam, 639 P.2d 858, 861 (Wash. Ct. App. 1982) (worked as prostitute while gathering evidence against brothel); GARY Y. MARX, UNDERCOVER: POLICE SURVEILLANCE IN AMERICA 236 n.9 (1988) [hereinafter MARX, POLICE SURVEILLANCE IN AMERICA] (purchased 10,000 illegally traded animals over eighteen months); Innocent Woman's ID Used by Police in Strip Club Sting, CINCINNATI POST, Apr. 11, 2005, at A9 (provided civilian agent with stolen identity to investigate strip club).

(7.) I define "democratic policing" as policing consistent with the rule of law and its associated values, such as accountability and transparency. This isn't the only way to define democratic policing, and in fact the term lacks a widely accepted meaning. For an exploration of the relationship between democratic theory and policing, see DAVID ALAN SKLANSKY, DEMOCRACY AND THE POLICE (2008).

(8.) See infra Part III.C.2.

(9.) Undercover policing operations involve both officers and civilian informants acting as secret agents. Because the involvement of undercover police officers in authorized criminality poses especially troubling legal, ethical, and normative issues, this Article will refer to police who work undercover, although much of the analysis could apply equally well to undercover informants. Those interested in the use of informants would do well to consult the work of Alexandra Natapoff, Snitching: The Institutional and Communal Consequences, 73 U. CIN. L. REV. 645 (2004).

(10.) A note on terminology: As will be discussed in Part II, authorized criminality may be considered legal for different reasons, either because it is justified by the public authority defense or because the officer lacks the required mental state of a potentially applicable crime. To simplify discussion in the text, I sometimes refer to police "participation in crime," although this does not mean that undercover officers are legally guilty of any crime.

(11.) There is no commonly used term for the practice. The FBI guidelines on undercover operations, described infra, refer to the practice as simply "otherwise illegal activity." See JOHN ASHCROFT, U.S. DEP'T OF JUSTICE, THE ATTORNEY GENERAL'S GUIDELINES ON FEDERAL BUREAU OF INVESTIGATION UNDERCOVER OPERATIONS (2002), available at 840c983e5800dd9cf0b6bd2349a5.pdf.

(12.) CHARLES BEENE, DECOY OPS: FIGHTING STREET CRIME UNDERCOVER 35-40 (1992) (describing activities of the San Francisco Street Crime Unit in which officers posed as citizen-victims).

(13.) Police, for instance, may stage an elaborate fictitious criminal organization in order to recruit a suspect, and subsequently bully him into confessing to his suspected crime on the ground that the organization's "boss" requires total honesty. See, e.g., R. v. Mentuck, [2001] 3 S.C.R. 442, 2001 SCC 76 (Can.) (discussing such a technique as part of "Operation Decisive" conducted by the Royal Canadian Mounted Police).

(14.) WAYNE LAFAVE, SUBSTANTIVE CRIMINAL LAW [section] 10.7(d) (2d ed. 2008) (noting that "excessive zeal in law enforcement rarely leads to a criminal prosecution of the police").

(15.) Cf. George I. Miller, Observations on Police Undercover Work, 25 CRIMINOLOGY 27, 27 (1987) ("Undercover work is arguably the most problematic form of policing undertaken by municipal police departments and [yet] little is known about it in operation.").

(16.) See Ross, supra note 6, at 576 ("An agent may 'violate the literal terms of certain penal statutes.... [But] at some point feigned participation in a crime bears such resemblance to the crime itself that society cannot tolerate the conduct.'" (quoting 2 G. ROBERT BLAKEY, TECHNIQUES IN THE INVESTIGATION AND PROSECUTION OF ORGANIZED CRIME: MANUALS OF LAW AND PROCEDURE 41 n.217 (1980))).

(17.) See, e.g., Alan Feuer, Bronx Man Pleads Guilty in Terror Case, N.Y. TIMES, Apr. 5, 2007, at BI (describing undercover police officer's role in investigation leading to guilty plea of "would-be" Al Qaeda operative Tarik Shah); William K. Rashburn, Detective Was "Walking Camera' Among City Muslims, He Testifies, N.Y. TIMES, May 19, 2006, at B1 (describing work of N.Y.P.D. detective recruited from police academy to play deep cover role in Brooklyn Muslim community to investigate Islamic extremists and testify at trial of Shahawar Siraj).

(18.) United States v. Kaminski, 703 F.2d 1004, 1010 (7th Cir. 1983). See also United States v. Jannotti, 673 F.2d 578, 616 (3d Cir. 1982), where the majority and dissenting opinions, while in violent disagreement about whether the facts supported a finding of entrapment as a matter of law, agreed that undercover officers could participate in illegal activity.

(19.) I have suggested elsewhere that this focus has also steered legal scholars away from paying close attention to private policing, which is not regulated by constitutional criminal procedure at all, but has become increasingly significant as a source of private crime control and order maintenance. See, e.g., Elizabeth E. Joh, The Paradox of Private Policing, 95 J. CRIM. L. & CRIMINOLOGY 49 (2004); see also David Sklansky, The Private Police, 46 UCLA L. REV. 1165 (1999); discussion infra Part IV.C.

(20.) One of the most important studies in this area is Gary T. Marx's Undercover: Police Surveillance in America, which provides a sociological analysis of undercover policing in the United States. See MARX, POLICE SURVEILLANCE IN AMERICA, supra note 6. Michel Girodo, cited infra in notes 65 and 217, has also conducted extensive psychological research on undercover police.

(21.) There are notable exceptions. Jacqueline Ross has written extensive comparative scholarship on undercover policing in the United States and in Western Europe. See, e.g., Ross, supra note 6.

(22.) Here is one imprecise measure: A Westlaw search in the "Journals and Law Reviews (JLR)" database for those articles with either the Fourth Amendment or Fifth Amendment in the title and with at least one mention of police in the text yields more than 1600 hits (1667 to be precise). A search in the same database of articles with "undercover" in the title and at least one mention of police in the text yields only thirty-five hits (a result which includes a heavy focus on the entrapment doctrine) (search conducted Oct. 2009).

(23.) Gary T. Marx, When the Guards Guard Themselves: Undercover Tactics Turned Inward, 2 POLICING & SOC'Y 151, 152 (1992).

(24.) Ross, supra note 6, at 571.

(25.) Once the Sixth Amendment's right to counsel applies to the defendant, however, the government may not elicit statements from the defendant in any manner, including through an undercover agent, without his attorney present. See Massiah v. United States, 377 U.S. 201, 201 (1964).

(26.) This is the Fourth Amendment's "third party doctrine," which is discussed infra text accompanying note 34.

(27.) See Jerome H. Skolnick, Deception by Police, CRIM. JUST. ETHICS, Summer-Fall 1982, at 40, 41 ("Deception occurs at three stages of the detecting process: investigation, interrogation, and testimony.").

(28.) Julius Wachtel, From Morals to Practice: Dilemmas of Control in Undercover Policing, 18 CRIME L. & SOC. CHANGE 137, 139 (1992).

(29.) Id. at 140-41.

(30.) See id. at 143. In bait-sales, officers pretend to be thieves with stolen goods available for sale to pawnshop owners and other potentially disreputable business persons. Street crime decoys will, for instance, frequent areas where pickpocketing is common and pose as easy targets. See generally CHARLES BEENE. DECOY OPS: FIGHTING STREET CRIME UNDERCOVER (1992).

(31.) See, e.g., Wachtel, supra note 28, at 144 ("Abolishing undercover work would make it impossible for the police to detect and investigate secretive and consensual crime.").

(32.) See, e.g., LUCAS A. POWE, JR., THE WARREN COURT AND AMERICAN POLITICS 379-444 (2000) (discussing dramatic changes in the recognization of criminal defendants' constitutional rights during the Warren Court era).

(33.) Cf. Gary T. Marx, The Interweaving of Public and Private Police in Undercover Work, in PRIVATE POLICING 172, 184-86 (Clifford D. Shearing & Phillip C. Stenning eds., 1987). Marx uses this argument to suggest that increasing restrictions on the public police will result in greater reliance upon private police, but it applies equally to investigative techniques used by public police that have been given varying levels of attention by the courts, particularly the United States Supreme Court.

(34.) See Hoffa v. United States, 385 U.S. 293 (1966); Lewis v. United States, 385 U.S. 206 (1966); Lopez v. United States, 373 U.S. 427 (1963). For a provocative defense of the often-criticized doctrine, see Orin S. Kerr, The Case .for the Third-Party Doctrine, 107 MICH. L. REV. 561 (2009).

(35.) Illinois v. Perkins, 496 U.S. 292 (1990).

(36.) See, e.g., Richard A. Leo, From Coercion to Deception." The Changing Nature of Police Interrogation in America, 18 CRIME L. 8,: SOC. CHANGE 35, 35 (1992). While it is true that 9/11 raised serious debate about the propriety of torture as an interrogation technique in terrorism investigations, the larger historical trend of policing generally has been toward less violent tactics.

(37.) United States v. Wick, 948 F.2d 1293 (9th Cir. 1991) (unpublished disposition).

(38.) United States v. Myers, 692 F.2d 823 (2d Cir. 1982); United States v. Murphy, 642 F.2d 699 (2d Cir. 1980); United States v. Myers, 635 F.2d 932 (2d Cir. 1980).

(39.) United States v. Parisi, 674 F.2d 126 (1st Cir. 1982).

(40.) MARX, POLICE SURVEILLANCE IN AMERICA, supra note 6, at 7.

(41.) James McKinley, Dogfighting Ring Is Broken Up in Texas, N.Y. TIMES, Nov. 16, 2008, at A26 (describing seventeen-month undercover operation in which Texas state police infiltrated an invitation-only underground dogfighting network); see also James McKinley, Dogfighting Subculture Is Taking Hold in Texas, N.Y. TIMES, Dec. 7, 2008, at A30 (describing "murky and dangerous subculture" involving illegal dogfighting, fencing stolen property, and illegal drugs).

(42.) See, e.g., EDWIN M. SCHUR, CRIMES WITHOUT VICTIMS: DEVIANT BEHAVIOR AND PUBLIC POLICY 80 (1965) ("The typical technique for effecting arrest involves the use of plainclothes detectives as decoys to draw indecent proposals....").

(43.) See, e.g., Wachtel, supra note 28, at 149; see also Skolnick, supra note 27, at 52 ("Practically speaking, it is impossible to enforce consensual crime statutes--bribery, drug dealing, prostitution--without employing deception.").

(44.) Miller, supra note 15, at 42.

(45.) MARX, POLICE SURVEILLANCE IN AMERICA, supra note 6, at 60-67.

(46.) See id. at 61-62.

(47.) See id. at 62-63.

(48.) Cf. id. at 62 (noting that entrapment is not an issue in postliminary surveillance investigations).

(49.) Id. at 64.

(50.) Henry W. Prunckun, Jr., It's Your Money They're After: Sting Operations in Consumer Fraud Investigations, 11 POLICE STUD.: INT'L REV. POLICE DEV. 190, 192 (1988) (describing such an effort by the South Australian Department of Public and Consumer Affairs); see also MARX, POLICE SURVEILLANCE IN AMERICA, supra note 6, at 65 (reporting similar tactic of U.S. Postal Service to deliberately distribute ads for weight loss or easy money schemes in order to identify potential fraud victims and contact and educate them).

(51.) See MARX, POLICE SURVEILLANCE IN AMERICA, supra note 6, at 65.

(52.) See id.

(53.) Marx describes these as the difference between "predicated" and "open ended" uses of undercover policing. See Marx, supra note 23, at 159.

(54.) See id. at 154.

(55.) For more discussion of the entrapment defense, see infra Part II.B.l.a.

(56.) Wachtel, supra note 28, at 143.

(57.) Or, as Wachtel puts it, such operations "run[] counter to the assumption that the Government should prevent rather than create crime." Id. at 149.

(58.) See MARX, POLICE SURVEILLANCE IN AMERICA, supra note 6, at 126-27.

(59.) See id.

(60.) For a compelling discussion of simulated environments and correspondence to real world behavior, see id.

(61.) See, e.g., Chancy v. Dep't of Law Enforcement, 393 N.E.2d 75, 76-78 (Ill. App. Ct. 1979) (describing undercover "Operation Suds" in which officers were asked to open and operate a tavern); MARX, POLICE SURVEILLANCE IN AMERICA, supra note 6, at 65 (describing example of officer posing as "a drunk with an exposed wallet").

(62.) See, e.g., D.W. Webster et al., Effects of Undercover Police Stings of Gun Dealers on the Supply of New Guns to Criminals, 12 INJ. PREVENTION 225, 225 (2006) (describing stings in Chicago; Detroit; and Gary, Indiana in which police posed as criminals attempting to buy guns illegally from targeted gun dealers).

(63.) United States v. Russell, 411 U.S. 423, 432 (1973); see also United States v. Parisi, 674 F.2d 126, 127 (1st Cir. 1982) (observing that because food stamps "are freely possessed and redeemed, and go through many hands, apprehension of parties whose possession or transfer is unlawful only for perhaps non-obvious reasons may be particularly difficult," and therefore it may be necessary "for the government itself to provide the stamps to the willing buyers").

(64.) Undercover police officers must sometimes use drugs to mask their identity, for example. Not all covert officers succumb to the pressure to use drugs, however, and some may successfully deploy other strategies to maintain their cover. As Bruce Jacobs reports from interviews with undercover officers, covert police sometimes evade drug use through the use of excuses (e.g., that they must report immediately to work after the score). In other cases, a covert officer will accuse the dealer of being a "narc" himself, or attempt to simulate drug use. See Bruce A. Jacobs, Undercover Drug-Use Evasion Tactics: Excuses and Neutralization, 15 SYMBOLIC INTERACTION 435 (1992). These alternatives, however, are likely to be most useful in "light cover" operations where the interaction between officer and criminal is fleeting.

(65.) Cf. Michel Girodo, Drug Corruption in Undercover Agents: Measuring the Risk, 9 BEHAV. SCI. & L. 361, 362 (1991) ("Criminals know to test the veracity of a proffered (false) identity by gauging the reaction to an invitation for drug use.").

(66.) Wachtel, supra note 28, at 146.

(67.) See, e.g., Michael Cooper, As Traps Grow, Wary Dealers Force Officers to Take Drugs, N.Y. TIMES, Feb. 17, 1997, [sections] 1, at 29.

(68.) State v. Rowan, 288 N.E.2d 829, 831 (Ohio Ct. App. 1972).

(69.) From his interviews with former heroin addicts and dealers, sociologist Bruce Jacobs also identified a sudden increase in the amount of drugs purchased and "transactional pushiness" (overeagerness to buy drugs) as additional tip-offs to criminals. See Bruce A. Jacobs, Undercover Deception Clues: A Case of Restrictive Deterrence, 31 CRIMINOLOGY 281, 286, 288 (1993).

(70.) See Bruce A. Jacobs, Getting Narced: Neutralization of Undercover Identity Discreditation, 14 DEVIANT BEHAV. 187 (1993) (discussing other tools needed for "mask maintenance").

(71.) See Jacobs, supra note 69, at 285.

(72.) For further discussion of the individual harms visited upon undercover investigators, see infra Part III.

(73.) See, e.g., Marx, supra note 23, at 163.

(74.) Light cover operations which pose the undercover investigator as a street crime decoy or drug buyer require much less professional investment than a deep cover operation in which the investigator may play a role for months, if not years, and undergo a "social death" of his real identity. See MARX, POLICE SURVEILLANCE IN AMERICA, supra note 6, at 85-86; Miller, supra note 15, at 28 (noting that the difference between the two is "the degree to which the officer's private life merges with the fictitious civic identity").

(75.) Cf. Marx, supra note 23, at 159 (1992) (discussing these factors as influencing the degree of internal controls in organizations).

(76.) See Ross, supra note 6, at 575 ("American law is, by and large, unwilling to use criminal sanctions to regulate and restrain undercover policing.").

(77.) Jacqueline E. Ross, The Place of Covert Surveillance in Democratic Societies: A Comparative Study of the United States and Germany, 55 AM. J. COMP. L. 493, 540 (2007).

(78.) See Ross, supra note 6, at 581 (observing that criminal sanctions play a "subordinate role in regulating undercover policing" as compared to internal guidelines, ethical rules for prosecutors, and defendants' use of the entrapment defense).

(79.) See LAFAVE, supra note 14, [sections] 10.7(d) (noting that in some cases, the participation of a police officer will not be criminal because a required mental state such as "feloniously" is absent).

(80.) See, e.g., UNIFORM CONTROLLED SUBSTANCES ACT [sections] 401(a) (1994) ("[A] person may not ... possess a controlled substances with intent to manufacture, distribute, or deliver, a controlled substance."). This is also true of federal law. See, e.g., 21 U.S.C. [sections] 841(a) (2006).

(81.) See, e.g., Ross, supra note 6, at 575. The common law defense has been recognized by statute in some states. See, e.g., IOWA CODE ANN. [sections] 704.11 (West 2009) ("[A] peace officer or person acting as an agent of or directed by any police agency who participates in the commission of a crime by another person solely for the purpose of gathering evidence leading to the prosecution of such other person shall not be guilty of that crime or of the crime of solicitation....").

(82.) See 2 PAUL H. ROBINSON, CRIMINAL LAW DEFENSES [sections] 142(a) (1984).

(83.) See id. [sections] 142(b). Many cases involving the defense involve considerations, usually ruling against the defendant, of whether the conduct in question was in fact authorized. See, e.g., Lilly v. West Virginia, 29 F.2d 61, 64 (4th Cir. 1928) (holding instructions on defense should have been given when federal prohibition agent struck and killed bystander while in pursuit of suspect); People v. Lesslie, 24 P.3d 22, 25 (Colo. Ct. App. 2000) (noting that defense does not apply when superior could not have legally authorized action in question); People v. Roberts, 601 P.2d 654, 656 (Colo. Ct. App. 1979) (rejecting defense where prison guard was not authorized to engage in undercover activity); Walker v. Commonwealth, 127 S.W.3d 596, 604-05 (Ky. 2004) (rejecting defense where defendant bail bondsman received no authorization to detain third party).

(84.) See ROBINSON, supra note 82, [sections] 142(d).

(85.) See id. [sections] 142(e) ("Despite the typical absence of an explicit general statement of the proportionality requirement, it would seem difficult to deny that proportionality should be a limitation on the law enforcement justification in every case."). In the context of deadly force, the Supreme Court has required a proportionality determination as a matter of federal constitutional law. Tennessee v. Garner, 471 U.S. 1 (1985).

(86.) See ROBINSON, supra note 82, [sections] 142(d).

(87.) See id. [sections] 142(e).

(88.) See LAFAVE, supra note 14, [sections] 10.7(d) ("[I]t would appear that in certain other circumstances [including undercover operations] the otherwise criminal conduct of a police officer, or a private person acting on behalf of an officer, may be privileged because the person was pursuing law enforcement purposes at the time." (citation omitted)); Ross, supra note 6, at 575 (discussing application of the defense to undercover operations). But see ROBINSON, supra note 82, [sections] 142 (omitting any mention of this variation of the defense).

(89.) See Ross, supra note 6, at 576; Ross, supra note 77, at 571 (noting that there is "no case-by-case assessment of what undercover conduct is 'really' criminal").

(90.) See infra Part III (discussing harms of authorized criminality).

(91.) Cf. Brizendine v. State, 627 S.W.2d 26, 28 (Ark. Ct. App. 1982) (explaining that while statutory public authority defense exists, "it is unlikely in the extreme that a policeman would be criminally prosecuted for such conduct"); LAFAVE, supra note 14, [sections] 10.7(d) (noting that the matter has "seldom been litigated or made the subject of legislation").

(92.) See Ross, supra note 77, at 540.

(93.) See, e.g., Chancy v. Dep't of Law Enforcement, 393 N.E.2d 75, 77 (Ill. App. Ct. 1979) (describing letters from U.S. Attorney and state assistant attorney general provided to undercover police "assuring [them] that they would not be prosecuted for their [undercover] activities" and declaring the undercover operation "lawful").

(94.) See, e.g., Ross, supra note 6, at 576 ("Although the imposition of criminal sanctions on covert agents remains a theoretical risk, it is not of practical importance in the day-to-day operation of the undercover policing system."); see also George E. Dix, Undercover Investigations and Police Rulemaking, 53 TEX. L. REV. 203, 284 (1975) ("A theoretically interesting problem is the criminal liability of undercover investigators who commit ... offenses."); id. at 286 ("[T]he formal law barely addresses this issue [of undercover officers' criminal liability], arguably depreciating the need to address the issue in law enforcement rulemaking.").

(95.) United States v. Nguyen, 413 F.3d 1170, 1177-81 (10th Cir. 2005) (concluding defendant was not entrapped as a matter of law).

(96.) See, e.g., State v. Bullock, 153 S.W.3d 882 (Mo. Ct. App. 2005).

(97.) Sorrells v. United States, 287 U.S. 435, 439-41 (1932).

(98.) See supra Part I.C.

(99.) See Skolnick, supra note 27, at 44.

(100.) See Kerr, supra note 34, at 591.

(101.) See, e.g., United States v. Russell, 411 U.S. 423, 436 (1973). A minority of jurisdictions apply an "objective" version of the doctrine, in which the defendant must show that the police conduct would have swayed a reasonable person. LAFAVE, supra note 14, [sections] 9.8(c). The Model Penal Code also has adopted an objective test. See MODEL PENAL CODE [sections] 2.13 (Proposed Official Draft 1962).

(102.) See LAFAVE, supra note 14, [sections] 9.8.

(103.) See, e.g., Bailey v. People, 630 P.2d 1062, 1065 n.5 (Colo. 1981) (en banc) (observing that in this view "the outcome varies with each individual defendant's state of mind; no general standards' governing the permissibility of police conduct are set" (emphasis added)).

(104.) Dix, supra note 94, at 276 (noting this observation about Russell, 411 U.S. at 423).

(105.) Although certainly in theory police involvement could be deemed so outrageous that it violates due process. See infra Part II.B. 1 .b.

(106.) See Ross, supra note 77, at 571 (observing that the "entrapment doctrine applies only to the most egregious pressures or temptations"); Skolnick, supra note 27, at 44 (noting that the subjective entrapment test "permits police to engage in deceptive practices provided that the deception catches a wolf rather than a lamb").


(108.) See, e.g., United States v. Jannotti, 673 F.2d 578, 604 (3d Cir. 1982) (en banc) (noting that there are "few" cases in which the defense has been successful); Ross, supra note 77, at 539 (observing that the "entrapment defense will matter to the outcome of criminal cases only in the most extreme and unusual cases"). Successful uses of the defense may also be rare because of the dramatic nature of the remedy; a successful entrapment defense operates as a bar to prosecution rather than as an exclusionary rule. See Dix, supra note 94, at 276.

(109.) And, of course, for entrapment to be raised at all, a suspect must become a defendant: another limitation of the defense. See Ross, supra note 6, at 590.

(110.) United States v. Citro, 842 F.2d 1149, 1152-53 (9th Cir. 1988) (implying that "outrageous government conduct" defense can apply even where defendant is predisposed to commit the offense).

(111.) Due process claims against the police are not limited to the undercover context. See, e.g., Rochin v. California, 342 U.S. 165, 172 (1952) (holding that involuntary stomach pumping of a defendant by police constituted a due process violation that "shocks the conscience").

(112.) See, e.g., People v. Isaacson, 378 N.E.2d 78, 83 (N.Y. 1978) (citing factors for state constitutional law due process including (1) whether the police manufactured the crime; (2) whether the police engaged in conduct "repugnant to a sense of justice"; (3) whether the defendant's reluctance to commit the crime is overcome by "appeals to humanitarian instincts such as sympathy or past friendship, by temptation of exorbitant gain, or by persistent solicitation in the face of unwillingness"; and (4) whether the facts reveal only a desire to obtain a conviction without a motivation to prevent further crime). Some jurisdictions suggest that the police conduct must be "malum in se or amount to the engineering and direction of the criminal enterprise from beginning to end." Citro, 842 F.2d at 1153 (citing United States v. Williams, 791 F.2d 1383, 1386 (9th Cir. 1986)).

(113.) See Shaw v. Winters, 796 F.2d 1124, 1125 (9th Cir. 1986) (holding that police conduct must be "repugnant to the American system of justice" (quoting United States v. Lomas, 706 F.2d 886, 891 (9th Cir. 1983))).

(114.) People v. Shine, 590 N.Y.S.2d 965,966 (App. Div. 1992) (mem.); cf. Robinson v. Cattaraugus County, 147 F.3d 153, 158-59 (2d Cir. 1998) (denying plaintiffs' motion for new trial on section 1983 claims involving similar police conduct); United States v. Smith, 924 F.2d 889, 897 (9th Cir. 1991) (noting that a defendant raising a due process claim against police "must meet an extremely high standard").

(115.) LaFave observes that the Supreme Court decisions in this area suggest that successful due process defenses "will be exceedingly rare." LAFAVE, supra note 14, [sections] 9.8(g); see also Hampton v. United States, 425 U.S. 484, 490-91 (1976) (undercover informant's supply of heroin subsequently sold by defendant to undercover police did not violate due process); United States v. Russell, 411 U.S. 423, 431-32 (1973) (undercover agent's supply to defendant of essential ingredient to manufacture methamphetamine did not violate due process).

(116.) In the Archer case, Judge Friendly made reference to such a hypothetical limit in dicta:
 [T]here is certainly a limit to allowing governmental involvement
 in crime. It would be unthinkable, for example, to permit
 government agents to instigate robberies and beatings merely to
 gather evidence to convict other members of a gang of hoodlums.
 Governmental 'investigation' involving participation in activities
 that result in injury to the rights of its citizens is a course
 that courts should be extremely reluctant to sanction.

United States v. Archer, 486 F.2d 670, 676-77 (2d Cir. 1973) (footnote omitted).

(117.) See State v. Williams, 623 So. 2d 462, 463 (Fla. 1993) (holding such conduct to violate the due process clause of the Florida state constitution).

(118.) See id. at 466 ("The delivery of a controlled substance in a reverse-sting operation is worlds apart from the manufacture of a dangerous controlled substance.").

(119.) See discussion of public" authority defense, supra Part II.A.2.

(120.) Indeed, it is not entirely clear that a due process defense exists in such a situation. See Hampton, 425 U.S. at 490 ("If the police engage in illegal activity in concert with a defendant beyond the scope of their duties the remedy lies, not in freeing the equally culpable defendant but in prosecuting the police under the applicable provisions of state or federal law.").

(121.) Thanks to Floyd Feeney for this observation.

(122.) See People v. Dlugash, 363 N.E.2d 1155, 1161 (N.Y. 1977) ("[T]he distinction between 'factual' and 'legal' impossibility was a nice one indeed and the courts tended to place a greater value on legal form than on any substantive danger the defendant's actions posed for society.").

(123.) Booth v. State, 398 P.2d 863,870-71 (Okla. Crim. App. 1964).

(124.) Id. at 867-68.

(125.) MODEL PENAL CODE [sections] 5.01(1) (Proposed Official Draft 1962).

(126.) MODEL PENAL CODE [sections] 5.01 cmt. (1985) ("The impossibility defense is rejected, liability being focused upon the circumstances as the actor believes them to be rather than as they actually exist."); see also People v. Thousand, 631 N.W.2d 694, 703 (Mich. 2001) (rejecting the impossibility defense where the defendant attempted to distribute obscene material to an undercover officer posing as minor online).

(127.) See, e.g., State v. Pacheco, 822 P.2d 183, 184 (Wash. 1994) (en banc) (reversing conspiracy conviction where defendant conspired with an undercover police officer).

(128.) MODEL PENAL CODE [sections] 5.03 explanatory note (1985) ("Guilt as a conspirator is measured by the situation as the actor views it....").

(129.) See, e.g., Paul H. Robinson & Markus D. Dubber, The American Model Penal Code: A Brief Overview, 10 NEW CRIM. L. REV. 319, 320 (2007).

(130.) See, e.g., Kyle S. Brodie, The Obviously Impossible Attempt: A Proposed Revision to the Model Penal Code, 15 N. ILL. U. L. REV. 237, 240 (1995) (noting the Model Penal Code's adoption of a "subjectivist viewpoint").

(131.) See Ross, supra note 77, at 539.

(132.) While other federal agencies--including the United Forest and Wildlife Service, the Internal Revenue Service, and the Drug Enforcement Agency--also have undercover guidelines, those used by the FBI are the most detailed. See Georg A. Wagner, United States' Policy Analysis on Undercover Operations, 9 INT'L J. POLICE SCI. & MGMT. 371, 373-74 (2007).

(133.) See ASHCROFT, supra note 11. The first Attorney General Guidelines for undercover operations were formalized in 1981, after intense public scrutiny of the FBI's involvement in the ABSCAM investigation. ABSCAM involved an FBI "sting" in which an informant posing as an agent for two fictitious Arab sheiks sought to influence a number of public officials. See GLENN A. FINE, U.S. DEP'T OF JUSTICE, THE FEDERAL BUREAU OF INVESTIGATION'S COMPLIANCE WITH THE ATTORNEY GENERAL'S INVESTIGATIVE GUIDELINES 41-42 (2005), available at

(134.) In September 2008, Attorney General Michael Mukasey announced new Guidelines for Domestic FBI Operations that attempt to harmonize guidelines previously considered inconsistent or ambiguous, as well as to emphasize the role of the FBI as an intelligence-gathering agency as well as a law enforcement agency. See Memorandum from Michael Mukasey, Attorney Gen., to the Heads of Dep't Components 2 (Sept. 28, 2008), available at (noting that while "[c]riminal law enforcement has always been central to the FBI's functions," "national security and intelligence aspects of its mission have increased in scope and importance since the September 11, 2001, terrorist attacks on the United States," and the new guidelines "integrate and harmonize standards"). While the new Guidelines for Domestic FBI Operations revise and repeal a number of then-existing guidelines, it leaves the Attorney General's Guidelines on Federal Bureau of Investigation Undercover Operations intact. See id. at 5.

(135.) ASHCROFT, supra note 11, at 3.

(136.) Id. at 8. Undercover operations which involve certain qualifying "fiscal circumstances," such as the purchase or lease of equipment, buildings, or facilities, also require approval from FBI Headquarters but need not be reviewed by the Undercover Operations Review Committee. Id. at 5-6.

(137.) Although the Guidelines refer to a $50,000 limit, in 2004 the FBI increased the limit to $100,000 in an internal electronic communication. Compare id. at 4, with FINE, supra note 133, at 148. Operations involving sensitive considerations may be extended for a period not to exceed thirty days. ASHCROFT, supra note 11, at 11.

(138.) ASHCROFT, supra note 11, at 12.

(139.) See entrapment discussion supra Part II.B.1.b.

(140.) ASHCROFT, supra note 11, at 12.

(141.) Id. at 13.

(142.) Id. at 12.

(143.) Id. at 13.

(144.) Id. at 12.

(145.) See Wagner, supra note 132, at 373 ("[Guidelines] can be created, changed or abolished at will, without notification or a comment period .... Agents found in violation of guidelines cannot be prosecuted for non-compliance.').

(146.) ASHCROFT, supra note 11, at 19.

(147.) BRIAN A. REAVES, U.S. DEP'T OF JUSTICE, CENSUS OF STATE AND LOCAL LAW ENFORCEMENT AGENCIES, 2004, at 1 (2007), available at The precise total, 777,885, includes fulltime and part-time sworn officers in local police forces, state troopers, sheriffs' offices, special jurisdiction police, and constables or marshals.


(149.) Henry Hamilton & John Ortiz Smykla, Guidelines for Police Undercover Work: New Questions About Accreditation and the Emphasis of Procedure over Authorization, 11 JUST. Q. 135, 145 (1994).

(150.) Id. at 147-48.

(151.) Id at. at 148.

(152.) Id. at 150.

(153.) See Ross, supra note 6, at 571.

(154.) Ross notes that this is a particularly American phenomenon. See id. at 576.

(155.) See id. at 574.

(156.) See id.

(157.) See id. at 588 (observing that an Italian undercover officer faces criminal prosecution if his conduct falls outside of statutory exemptions).

(158.) Id. at 574-75 (citing Legge di giugno, 1990, n. 162, Art. 25(1), Art. 97).

(159.) Id. at 574.

(160.) R. v. Campbell, [1999] 1 S.C.R. 565, 1999 SCC 676 (Can.), available at

(161.) See id.

(162.) See id.


(164.) An Act to Amend the Criminal Code (Organized Crime and Law Enforcement), 2001 S.C., ch. 32 (Can.). For a critical view of the Act, see Marc S. Gorbet, Bill C-24'S Police Immunity Provisions." Parliament's Unnecessary Legislative Response to Police Illegality in Undercover Operations, 9 CAN. CRIM. L. REV. 35 (2004) (arguing Section 25.1 is a redundancy and may actually inadvertently restrict undercover operations).

(165.) Canada Criminal Code, R.S.C., ch. C.-46, [section] 25.1(8) (1985). The provision also insulates from liability civilians who act under the authority of a police officer. See id. [section] 25.1(10).

(166.) Id. [section] 25.1(11)(1985).

(167.) Id. [section] 25.3(1) (1985). The annual reports require only limited information regarding occasions when the police have acted in ways that are covered by the justification. Details about investigations are not released.

(168.) See, e.g., JOHN KLEINIG, THE ETHICS OF POLICING 137 (1996) ("[T]he question has to be asked: What does [undercover policing] do for our sense of who we are and what our society represents that this or that kind of deception is sponsored by government officials.'?"); cf. MARX, POLICE SURVEILLANCE IN AMERICA, supra note 6, at 206 ("At best, in a democratic society, it will never be possible to be too enthusiastic about undercover operations."). Note too that some commentators defend undercover operations on the basis that these tactics target criminals beyond the reach of ordinary street policing. See, e.g., John Braithwaite et al., Covert Facilitation and Crime: Restoring Balance to the Entrapment Debate, 43 J. Soc. ISSUES 5, 19 (1987) (arguing that undercover policing "promote[s] greater equality between the treatment of the powerless and the powerful in the criminal justice system").

(169.) See, e.g., Hamilton & Smylka, supra note 149, at 136 ("Unlike measures of reactive [police] patrol.... undercover police work has no standard means of measurement, nor is there a central agency to receive the reports.").

(170.) 5 U.S.C. [section] 552(b)(7) (2006).


(172.) 5 U.S.C. [section] 552(b)(7)(E) (2006).

(173.) See FOIA GUIDE, supra note 171, at 813 ("[A]n ever growing body of case law demonstrates that this exemption applies to a very broad range of law enforcement information.... ").

(174.) Wickline v. FBI, No. 92-1189 SSH, 1994 WL 549756, at *5 (D.D.C. Sept. 30, 1994) (quoting Parker v. Dep't of Justice, No. 88-0760, slip op. at 8 (D.D.C. Feb. 28, 1990), aff'd, 934 F.2d 375 (D.C. Cir. 1991)).

(175.) See, e.g., Foster v. United States Dep't of Justice, 933 F. Supp. 687, 693 (E.D. Mich. 1996) (observing that disclosure of IRS undercover techniques "would diminish the effectiveness of the use of similar techniques in existing and future investigations"); Wagner v. FBI, No. 90-1314-LFO, 1991 U.S. Dist. LEXIS 7506, at *7 (D.D.C. June 4, 1991) (approving withholding of DEA "undercover techniques" as 7(E) exemption); FOIA GUIDE, supra note 171, at 818 n. 16 (citing these and other non-published cases regarding undercover police).

(176.) See, e.g., Hale v. United States Dep't of Justice, 973 F.2d 894, 902-03 (10th Cir. 1992), vacated and remanded, 509 U.S. 918 (1993), overruled on other grounds, 2 F.3d 1055 (10th Cir. 1993).

(177.) See, e.g., Maguire v. Mawn, No. 02Civ.2164(RJH)(MHD), 2004 WL 1124673, at *2 (S.D.N.Y. May 19, 2004).

(178.) See, e.g., id. at *3 (noting that use of FBI bait money "is particularly worthy of protection when the method employed is meant to operate clandestinely, unlike guards or bullet-proof glass barriers that serve their crime prevention purpose by operating in the open").

(179.) See Erik K. Luna, Transparent Policing, 85 IOWA L. REV. 1107, 1132, 1141 (2000).1

(180.) See id. at 1120. While Luna's primary concerns are instances of police abuse and corruption, his concerns about transparency have general applicability to all police actions that remain largely secret.

(181.) See id. at 1156.

(182.) Id. at 1160.

(183.) Id. at 1158-63.

(184.) For a discussion of police discretion and traffic stops, see Elizabeth E. Joh, Discretionless Policing: Technology and the Fourth Amendment, 95 CAL. L. REV. 199 (2007). Before the 1950s, scholars of the police assumed the police exercised hardly any discretion at all. For further discussion of discretion as a scholarly enterprise, see GEORGE L. KELLING, U.S. DEP'T OF JUSTICE, "BROKEN WINDOWS" AND POLICE DISCRETION 22 (1999); SAMUEL WALKER, TAMING THE SYSTEM: THE CONTROL OF DISCRETION IN CRIMINAL JUSTICE 6-12 (1993).

(185.) See, e.g., Luna, supra note 179, at 1133.

(186.) See Joh, supra note 184, at 207.

(187.) See id. at 207-08.

(188.) See JEROME H. SKOLNICK, JUSTICE WITHOUT TRIAL: LAW ENFORCEMENT IN DEMOCRATIC SOCIETY 69 (Macmillan 1994) (1966) ("If the central task of the administration of criminal law is to balance the conflicting principles of order and of legality, the dilemma is epitomized in the question of police discretion."); see also Albert Reiss, Jr., Police Organization in the Twentieth Century, 15 CRIME & JUST. 51, 74 (1992) ("Although the foundation of policing is the legal order and its rules, police officers, nevertheless, have enormous discretionary powers to apply the law.").

(189.) I refer to sociologist Max Weber's classic definition of the state in terms of its monopoly over the use of legitimate force. See MAX WEBER, POLITICS AS A VOCATION, reprinted in FROM MAX WEBER: ESSAYS IN SOCIOLOGY 77 (H.H. Gerth & C. Wright Mills eds. & trans., Routledge 2003) (1946).

(190.) See Joh, supra note 184, at 205.

(191.) See Colorado v. Bertine, 479 U.S. 367, 374 (1987).

(192.) Whren v. United States, 517 U.S. 806, 812-13 (1996).

(193.) See Atwater v. City of Lago Vista, 532 U.S. 318, 326, 354 (2001).

(194.) City of Chicago v. Morales, 527 U.S. 41, 64 (1999).

(195.) Whether identified as quality of life policing or broken windows policing, these strategies owe their empirical and normative foundations to the "broken windows" thesis of Kelling and Wilson: namely, that the failure of the police to enforce minor offenses invites the commission of more serious crimes. See James Q. Wilson & George L. Kelling, Broken Windows, ATLANTIC MONTHLY, Mar. 1982, at 29, 31-32.

(196.) William Kleinknecht, Perspective, Arresting Crime, STAR-LEDGER (Newark, N.J.), Feb. 27, 2000, at 1 (describing adoption of broken windows policing by major cities).

(197.) See, e.g., Dan M. Kahan & Tracey L. Meares, Foreword: The Coming Crisis of Criminal Procedure, 86 GEO. L.J. 1153, 1171-75 (1998) (arguing for a political process approach that mandates judicial deference to community preferences); Debra Livingston, Police Discretion and the Quality of Life in Public Places: Courts, Communities, and the New Policing, 97 COLUM. L. REV. 551, 650-70 (1997) (suggesting nonconstitutional restraints including department guidelines and community monitoring).

(198.) See, e.g., Dix, supra note 94, at 293 ("Undercover investigations' lack of a pervasive unifying doctrinal framework contrasts with other police practices, such as the traditional search of premises for evidence of criminal guilt."); Wagner, supra note 132, at 373 ("Compared with traditional police practices, undercover methods are relatively unhindered by constitutional or legislative restrictions.").

(199.) See, e.g., Ross, supra note 77, at 539 (observing that police have "discretion to decide when and whether to intervene, just as prosecutors have discretion not to bring charges").

(200.) See Miller, supra note 15, at 28.

(201.) Cf. Wagner, supra note 132, at 375 ("There are no clear legal limitations on the length of the operation, the intimacy of the relationships formed, the degree of deception used and the degree of temptation offered and the number of times it is offered. Police have much discretion with deciding on the outer limits of permissible undercover behaviour.").

(202.) See discussion supra Part II.A & B.

(203.) United States v. Murphy, 768 F.2d 1518, 1528 (7th Cir. 1985).

(204.) See United States v. Kaminski, 703 F.2d 1004, 1010 (7th Cir. 1983).

(205.) Murphy, 768 F.2d at 1528.

(206.) Id. at 1529.

(207.) Cf. United States v. Simpson, 813 F.2d 1462, 1468 (9th Cir. 1987) ("We recognize that many people in our society may find the deceptive use of sex in law enforcement to be morally offensive. Nonetheless, 'in order to apprehend those engaged in serious crime, government agents may lawfully use methods that are neither appealing nor moral if judged by abstract norms of decency.'" (quoting United States v. Bogart, 783 F.2d 1428, 1438 (9th Cir. 1986))).

(208.) As former Assistant Attorney General of the Criminal Division Philip Heymann stated before Congress: "Instead of having to rely on ... testimony of unsavory criminals and confidence men, . . . undercover techniques [permit us to] muster the testimony of credible law enforcement agents." FINE, supra note 133, at 138 (quoting FBI Undercover Guidelines: Oversight Hearing Be]ore the Subcomm. on Civil and Constitutional Rights o[" the H. Comm. on the Judiciary, 97th Cong. 130 (1981)); see also Wagner, supra note 132, at 372 (noting that in undercover operations "evidence can be presented in the most reliable form: direct testimony from agents who personally participated in unlawful conduct").

(209.) See FINE, supra note 133, at 41.

(210.) See id. at 42-43. The House of Representatives convened the House Subcommittee on Civil and Constitutional Rights, which began hearings in 1980 and issued a report in 1984. Id. at 42. The Senate's Select Committee to Study Undercover Activities issued a final report in 1982. Id. at 43.


(212.) Miller, supra note 15, at 35.

(213.) See, e.g., id. at 35 (reporting results of interviews with fourteen undercover officers in which eight working in deep-cover roles had received training of less than one day); see also Chancy v. Dep't of Law Enforcement, 393 N.E.2d 75, 76 (Ill. App. Ct. 1979) (noting that officers asked to operate a tavern for an undercover operation had no undercover experience or training).

(214.) Details about the detective's assignment were revealed during his testimony at the trial of Shahawar Matin Siraj, charged with plotting to blow up the Herald Square subway in 2004. Rashbaum, supra note 17, at B1.

(215.) See Girodo, supra note 65, at 365, 369 (reporting these findings from a survey of 271 undercover federal agents from unnamed department); see also DiGloria v. Chief of Police, 395 N.E.2d 1297, 1299 (Mass. App. Ct. 1979) (describing officer's heroin addiction that began during undercover assignment to infiltrate illegal drug trade).

(216.) Mark R. Pogrebin & Eric D. Poole, Vice Isn't Nice: A Look at the Effects of Working Undercover, 21 J. CRIM. JUST. 383, 389-91 (1993) (reporting results from qualitative interviews with undercover investigators).

(217.) See Michel Girodo, Undercover Agent Assessment Centers: Crafting Vice and Virtue for Impostors, 12 J. SOC. BEHAV. & PERSONALITY 237, 243 (1997) [hereinafter Girodo, Assessment Centers]; see also Michel Girodo et al., Dissociative-Type Identity Disturbances in Undercover Agents: Socio-Cognitive Factors Behind False-Identity Appearances and Reenactments, 30 J. Soc. BEHAV. & PERSONALITY 631, 631 (2002) (documenting evidence in which undercover agents in training exercises reported and were observed as involuntarily manifesting alternate identities outside of investigation context); Michel Girodo, Symptomatic Reactions to Undercover Work, 179 J. NERVOUS & MENTAL DISEASE 626, 628 (1991) [hereinafter Girodo, Symptomatic Reactions] (documenting survey evidence suggesting that undercover work related to higher-than-average incidence of psychiatric problems).

(218.) See Marx, Under-the-Covers, supra note 6, at 23 n.34 (discussing perils of agents who "go[] native" and fall in love with targets); see also Miller, supra note 15, at 40 ("[I]ncidents where former undercover officers are later found to have become violators themselves are common.... " (citations omitted)).

(219.) See Girodo, Assessment Centers, supra note 217, at 238 (noting that undercover "work selection criteria" includes "personality predispositions of both integrity and deceit").

(220.) Cf. Michel Girodo, Health and Legal Issues in Undercover Narcotics Investigations. Misrepresented Evidence, 3 BEHAV. SCI. & LAW 299, 307 (1985) (observing that law enforcement agencies often seek such persons as candidates for undercover assignments without realizing accompanying risks).

(221.) Miller, supra note 15, at 32-33.

(222.) See Pogrebin & Poole, supra note 216, at 389.

(223.) See id.; see also Miller, supra note 15, at 40 ("Citizens are not only candidates for arrest, they are social companions, confidants to some extent, and perhaps lovers.").

(224.) See Marx, supra note 23, at 160.

(225.) See Pogrebin & Poole, supra note 216, at 391,393 ("The norms of police ethics may thus be turned upside down in undercover work.").

(226.) See discussion supra Part II.A.

(227.) See Girodo, Assessment Centers, supra note 217, at 244.

(228.) See Marx, supra note 23, at 162.

(229.) See id. at 163.

(230.) KLEINIG, supra note 168, at 137.

(231.) Wachtel, supra note 28, at 144.

(232.) Cf. Marx, supra note 23, at 167 ("When [the police] are viewed as moral exemplars and beyond rapproach, there is probably less violation of the rules they are charged with enforcing and public cooperation is greater."); Wachtel, supra note 28, at 139 ("Government lying promotes cynicism and can break the bonds of trust that give representative government its special appeal.... "(citation omitted)).

(233.) Olmstead v. United States, 277 U.S. 438, 485 (1928) (Brandeis, J., dissenting).

(234.) Alexandra Natapoff makes a similar case for the expressive norms conveyed by the pervasive use of criminal informants. See Natapoff, supra note 9, at 682-83.

(235.) For current national data on arrests and clearances, see SOURCEBOOK OF CRIMINAL JUSTICE STATISTICS ONLINE (2009), [hereinafter SOURCEBOOK ONLINE].

(236.) See, e.g., Hamilton & Smykla, supra note 149, at 136 ("Our knowledge about undercover police work today is slight in comparison with our knowledge of reactive law enforcement."). Of course, undercover activity is not the only area in which greater empirical data collection is needed. With respect to police shootings, there is insufficient information to compile an accurate picture of how many times people have been killed by the police and when they use excessive force. This is true despite the fact that federal law requires the Attorney General to collect this data and publish it annually. See Fox Butterfield, Bookkeeping: When the Police Shoot, Who's Counting?, N.Y. TIMES, Apr. 29, 2001, at WK5.

(237.) Cf. Ross, supra note 6, at 585 ("No branch or agency of the U.S. government systematically considers the price we pay for allowing undercover operations to infiltrate and potentially distort a variety of social, economic and political settings in which licit and illegal activities coexist.").

(238.) See supra Part III.A.


(240.) The Sourcebook of Criminal Justice Statistics publishes information on arrest data including the offense charged; the age, race, and sex of persons arrested; and geographic region. See SOURCEBOOK ONLINE, supra note 235.

(241.) Thus, a public reporting requirement similar to the one that exists under Canadian law does not go far enough. While Canadian law requires an annual report when its law enforcement defense is invoked, no other details of the investigation are provided. See supra Part II.D.

(242.) Dix, supra note 94, at 208.

(243.) See discussion supra Part II.C.

(244.) See KELLING, supra note 184, at 17.

(245.) See id. at 15.

(246.) Livingston, supra note 197, at 663.

(247.) See Braithwaite et al., supra note 168, at 11. Braithwaite and his co-authors express general skepticism that guidelines can "provide much practical protection against [police] abuse." Id. at 12.

(248.) See David A. Sklansky, Quasi-Affirmative Rights in Constitutional Criminal Procedure, 88 VA. L. REV. 1229, 1272-73 (2002) (discussing "broad consensus" that had developed by the 1970s about the desirability of guidelines to limit police discretion).

(249.) See KELLING, supra note 184, at 28.

(250.) See id. at 16 (observing that the New York City Transit Police Department in 1980s was "virtual[ly] silen[t]" about practical police work); Samuel Walker, Controlling the Cops: A Legislative Approach to Police Rulemaking, 63 U. DET. L. REV. 361, 368 (1986).

(251.) See, e.g., Sklansky, supra note 248, at 1273.

(252.) See, e.g., Livingston, supra note 197, at 662 (observing that legislatures have not generally forced police rulemaking out of concern that they appear "anti-police").

(253.) See supra note 22.

(254.) State and federal laws play an important, but much less significant role in these areas, and have had the most sway where the Court has failed or declined to have significant authority, such as the regulation of government records, phone numbers, email, and other records. See William J. Stuntz, The Political Constitution of Criminal Justice, 119 HARV. L. REV. 781, 789 (2006). But see Orin S. Kerr, The Fourth Amendment and New Technologies: Constitutional Myths and the Case for Caution, 102 MICH. L. REV. 801, 850-57 (2004).

(255.) See Stuntz, supra note 254, at 782.

(256.) See id. at 785.

(257.) See id. at 790; see also id. at 810 ("Constitutional law made governing policing hard, governing litigation somewhat easier, and governing punishment very easy indeed. Legislators have spent accordingly.").

(258.) See id. at 803-04.

(259.) Cf. id. at 835 (arguing that police corruption should be a "subject of constitutional concern").

(260.) Austin Sarat & Susan Silbey, The Pull of the Policy Audience, 10 LAW & POL'Y 97 (1988).

(261.) See id. at 99, 141.

(262.) See id. at 131 ("Research which addresses the policy elite ... speaks with a particular voice.... In particular, the pull of the policy audience leads sociologists of law to ignore perspectives inconsistent with its epistemology, or purposes.").

Elizabeth E. Joh, Visiting Professor of Law, Stanford Law School; Professor of Law, University of California, at Davis ( Thanks to Mariano-Florentino Cuellar, Floyd Feeney, George Fisher, Gary Marx, Charles Reichmann, Robert Weisberg, the participants in the 2008 Hixon-Riggs Forum for Science and Technology, the U.C. Davis Law Faculty Workshop and the Stanford Law Faculty Workshop for their helpful comments and conversations; to the staff of the Stanford Law Review for their editorial assistance; to the staff of the Mabie Law Library at U.C. Davis School of Law, Erin Murphy, Jacob Storms, and Andrew Zee for their research assistance; to former Dean Rex Perschbacher and current Dean Kevin Johnson of the U.C. Davis School of Law for their support.
COPYRIGHT 2009 Stanford Law School
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2009 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Joh, Elizabeth E.
Publication:Stanford Law Review
Date:Dec 1, 2009
Previous Article:Delaware's shrinking half-life.
Next Article:Nonlethal self-defense, (almost entirely) nonlethal weapons, and the rights to keep and bear arms and defend life.

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |