Desktop version

Home arrow Political science

Integrity Tests

Many undercover operations might be considered to be entrapment in the ordinary common sense meaning of that term, i.e. to trap someone. (This sense of entrapment is to be distinguished from various legal definitions of the term.) Consider, for example, the NSW Royal Commission operation in which police officer Trevor Haken trapped or entrapped police officer Chook Fowler (Miller and Blackler 2005, Chap. 5).1 Clearly the infringement of the right to privacy is a central feature of undercover operations in which a police officer establishes a relationship and gains the trust of an offender. Indeed, important questions arise here as to the morally admissible nature and extent of such relationships. It is one thing to establish friendly relations, it is another to establish a sexual relationship. In some undercover operations, police in effect act as observers, albeit inside-observers. The offenders commit the offences that they commit independently of the actions of the undercover operatives. However, often undercover operatives interact with offenders in such a way as to make a difference to whether or not, or when, where or how, an offence is committed. This was the case with Chook Fowler; Haken was not simply an observer, he was also an active participant. This is trapping in our target sense of the term.

Integrity testing is a form of trapping used by internal affairs departments in their efforts to stem police corruption. Integrity testing is a pro-active law enforcement strategy used in many jurisdictions in preference to—or at least, in addition to— reactive strategies, such as complaints investigations. Integrity testing can make use of undercover operatives posing as drug buyers and the like. It can involve the building of lengthy interpersonal relationships. The most important consideration in favour of trapping is in relation to serious systemic police corruption. For example, in the 1990s corruption in the NSW Police was systemic, and evidently the only way to bring corrupt police to justice was by way of trapping corrupt officers. This typically involved “turned” corrupt police officers operating undercover. Only such officers would be trusted by corrupt fellow officers, and only a managed scenario would enable reliable evidence, such as videotapes, to be obtained.

Integrity testing can be random or targeted. Targeted integrity testing focuses on a specific person (or persons) who is/are reasonably believed to be involved in crime. Random integrity testing is not directed at any specific person. Integrity testing raises a number of ethical issues, including: (a) deception; (b) the infringement of privacy; (c) uncertainty in relation to the moral culpability of the offender, i.e. the offender was “tricked” into doing what he or she otherwise would not have done, and; (d) impropriety of law enforcement agents, since they might be creating crimes that otherwise would not exist.

Accordingly, questions arise as to the moral and legal limits that ought to be placed on integrity testing. The options here range from banning all forms of integrity testing, to allowing certain kinds of integrity testing in relation to a


An earlier version of the material in this section appeared in Miller and Blackler (2005, Chap. 5).

narrowly circumscribed set of crimes. Here there are at least two relevant preliminary considerations. First, many serious crimes, such as murder, rape, and grievous bodily harm, do not lend themselves to integrity testing. After all, integrity testing must involve the actual commission of a ‘crime’, and presumably allowing someone to be murdered in order to convict the murderer is morally unacceptable. On the other hand, some related crimes, such as conspiracy to commit murder, might be suitable for integrity testing. Second, given the morally problematic nature of integrity testing, arguably it should only be used sparingly and presumably only for serious crimes—at least in relation to ordinary citizens. On the other hand, arguably police officers are a difficult kettle of fish; arguably, police need to meet higher moral standards than ordinary citizens, at least in certain respects. So perhaps random or targeted integrity testing of thieves among the ordinary citizenry is morally unjustified, but integrity testing of thieves within the ranks of police officers is justified under certain conditions.

Let us briefly consider deception in relation to integrity testing. If a suspect is to be trapped, he or she will need to be deceived. However, such deception will occur at the investigatory stage of police work. Evidently, when deception occurs at the investigatory stage—as opposed to the testimonial stage—it may well be morally justifiable. Thus lying to a corrupt officer to enable an arrest is morally justified, whereas lying in court is not morally justified. On the other hand, it is normally unlawful to give false information to a police officer; and this for good reason, given the need for police to elicit the truth from witnesses and others if crimes are to be solved.

Apparently, the illegality of giving false information to a police officer has hindered integrity testing by some internal affairs departments in some jurisdictions.2 For this provision limits the range of possible trapping scenarios. Scenarios in which, for example, drugs or money are left lying around are admissible, but not ones in which undercover police give false information to corrupt officers. A related problem for integrity testing is the ability of undercover police to assume a false identity. Under legislation in some jurisdictions, such as the Crimes (Assumed Identities) Act 2004 in Victoria, Australia, police can assume a false identity. However, evidently there have been practical problems in that agencies, e.g. the driving licence issuing authority, are not able to be compelled to comply with requests for false identity documentation and/or they are not able—or not able except with great difficulty—to change their databases so as to indicate that the licenses issued to these false identities are longstanding and not simply recently issued. These practical problems arise in the context of corrupt police officers having access to the databases in question and being suspicious of any allegedly substantiating documentation with respect to identity. Evidently, these practical problems can be resolved by amendments to the relevant legislation; the legislation can require such government authorities to comply with police requests. A related issue here is the need for information technology infrastructure that is resistant to


For example, the Ethical Standards Department (ESD) of Victoria Police in Australia in the past.

abuse and misuse on the part of police, i.e. infrastructure that either prevents, or provides precise audit trails of, officers’ unauthorised access to data of the kind in question.

Let us now turn to privacy issues. Infringement of privacy in integrity testing is morally justifiable under certain conditions. Privacy is not an absolute right, whether privacy on the telephone, the car phone, the Internet or on any other communication or information system. The rights to privacy of some individuals including police officers, and the right to confidentiality of members of some organisations including police organisations, will in some cases be overridden by the rights of other individuals, other members of organisations and the community in general to be protected by the law enforcement agencies from the perpetrators of serious crimes, including corrupt police.

Infringements of privacy, including the privacy of police officers, by law enforcement officials are morally justifiable if certain conditions are met. These conditions include the following ones: (a) there is reasonable suspicion that the person whose privacy is to be infringed intends to commit a serious crime; (b) the methods in question are effective, and; (c) there is no alternative non-intrusive, or less intrusive, method of investigation.

Arguably, integrity testing is required—or is far more effective than reactive methods, such as investigating complaints—in relation to certain crimes. The crimes in question include ones that do not involve a complainant, e.g. drug-dealing, or are in areas such as crime within police organisations, where offences might be difficult to prove because offenders are themselves experts in criminal investigation and protected by a secretive and fraternal culture. Moreover, in relation to certain kinds of offence and offender, arguably integrity testing does better on a cost/benefit analysis than reliance on informers, or on undercover operatives who observe but do not trap. Informers often provide unreliable information, and often fail to provide evidence of the guilt of those they implicate in crimes. Undercover operations are resource intensive and their outcomes uncertain. This is especially so when undercover operatives simply wait for a suspect to create the opportunity to commit a crime, and then hope to gather evidence in relation to the crime when it does happen. By contrast, integrity testing involves stage-managing a crime at a time and place chosen by police; so there is a greater assurance that the crime will be recorded and the offender convicted.

If persons who have been trapped are justifiably to be convicted, then they must have committed a crime. However, even if they have performed a criminal act, there might be important reasons not to convict them. Specifically, they might have been the victims of morally unjustified entrapment. What tests ought to be applied to determine whether someone was the victim of morally unjustified entrapment? In Australia the Crimes (Controlled Operations) Act 2004 at s.14 (d) states: “that the operation will not be conducted in such a way that a person is likely to be induced to commit an offence against a law of any jurisdiction of the Commonwealth that the person would not otherwise have intended to commit”. This introduces the idea of an offence that would not otherwise have been committed, i.e. the creation of crime, and the related idea of an excessive inducement. In the USA, two legal tests to determine whether someone has been entrapped have been proposed; the subjective test and the objective test. However, only the subjective test is actually in used. Note that in the sense of “entrapment” in question in the legal environment of the USA, entrapment is necessarily unlawful; in the USA, entrapment, by definition, involves pro-active policing practices that fail (in particular) the subjective test.

The subjective test asks whether the suspect has a disposition to commit crimes of the kind in question. Theoretically, but not necessarily, or indeed actually, in law, we might establish the existence of a disposition on the basis of his/her past behaviour, e.g. past criminal convictions. Evidently the point of this test is to ensure that the person entrapped has the requisite degree of culpability; an important motivating reason for using this test is the concern that without it, the police might induce an intention or inclination to commit a crime that was otherwise absent.

The objective test asks whether or not the State has acted improperly by virtue of instigating the crime. This resolves itself into two issues. The first issue is whether or not the contribution of the police to the creation of the opportunity to commit the crime is excessive. For example, suppose an undercover police officer supplies a fellow officer with the raw materials and the equipment to manufacture heroin, and suppose that the raw materials and equipment is not available to the person from any other source(s). The second issue is whether or not the inducement offered to commit the crime was unreasonable (too strong), e.g. offering a junior police officer working in a low risk area a million dollars to engage in minor theft.

One problem for the subjective test is how to provide evidence of a disposition. This problem is heightened in legal contexts in which knowledge of past crimes and convictions is not normally allowed to be used in determining guilt in relation to a current crime. A further possible problem for the subjective test is that it does not rule out strong inducements. Police officers might abuse the system by offering inducements that are too strong, and yet conviction would follow if the suspects had strong dispositions to commit the crime. A related problem arises from the fact that a disposition to commit a crime is not equivalent to an intention to commit that crime. Suppose someone has a disposition to commit a crime. However, knowing that he has this disposition, he puts himself in a context in which there is no opportunity to commit the crime. Consider a police officer who has a cocaine habit but who wants to avoid taking cocaine and decides to avoid going to places in which cocaine is available. Now assume an undercover police officer gets the police officer in question drunk and pressures him to join him in using cocaine. Arguably, the mere presence of a disposition is not sufficient for morally justified trapping; so the subjective test—at least as described above—would have to be strengthened.

A possible problem for the objective test is that it protects some people who should be found guilty. Suppose strong inducements are used in cases of suspects with strong dispositions to commit the crime, and suppose these suspects are in fact guilty of this kind of crime. Such inducements will be ruled out by the objective test, and yet the guilty persons in question will go free. On the other hand, it is normally preferable that some of the guilty go free than that some of the innocent are convicted. So this objection is relatively weak.

Another objection is that the objective test—in so far as it involves random testing of ordinary citizens, as opposed to police officers—amounts to the government engaging in integrity testing of its citizens. This is surely unacceptable; governments have no right to convict a citizen merely because the citizen fails to resist an inducement to commit a crime, even if it is an inducement that they ought to have resisted. However, it is by no means clear that this objection stands in the case of random integrity testing of police officers. A stronger objection to the objective test is that it is not a particularly effective test of virtue. For someone who lacked the disposition to commit that kind of crime, or indeed crimes in general, might nevertheless fail the objective test on a single occasion.

What is surely acceptable in the case of police officers is targeted integrity testing of individuals reasonably suspected of committing the crime that is the subject of the test. Moreover, random integrity testing of certain categories of public servants, such as police or politicians, in relation to a circumscribed set of crimes might be acceptable under certain conditions. For example, suppose bribe-taking is rife in a specific police organization and is undermining police operations in relation to serious criminality, and all other measures have failed to curtail it; perhaps random integrity testing is now warranted. The general moral justification for this is that police (and some other government officials) need to have a certain standard of integrity in relation to specific kinds of inducement, and they voluntarily accept a public office on the basis that they meet that standard. Accordingly, their integrity might reasonably be open to testing, especially if it is made clear to them before they accept the public office that their integrity might be subjected to a test.

There is a general objection to integrity testing, and this objection apparently stands irrespective of whether the subjective test or the objective test is applied. This is the objection that integrity testing involves the creation of crime, rather than the detecting or preventing of crime that would have existed independently of trapping. If this objection is sustained, it is decisive; integrity testing should be abandoned. But is this objection sustained?

In order to assist our deliberations, consider the following. Suppose a desk-bound police officer, Officer A, forms an intention to commit the one-off crime of stealing $5000 of drug money. Officer A believes the money was abandoned by his drug-dealer neighbour, B, in the garden outside B’s house when B was arrested by the police, and that his crime will go undetected. Suppose that, unknown to A, this money was in fact confiscated by the police. However, the police decide not to remove the money, but rather to leave it with the purpose of trapping A, who they suspect might be tempted by the prospect of such “easy money”, notwithstanding his general compliance with the law. A goes to steal the money and is caught red-handed. Notice that if the objective test is applied, the police are entitled to engage in this kind of trap. In the first place, the inducement, viz. $5000, is of a kind that police officers could reasonably be expected to resist. In the second place, it was the drug-dealer who created the opportunity for theft; all the police did was fail to remove this opportunity. On the other hand, this kind of integrity test is ruled out by the subjective test; for A does not have a disposition to steal.

Given the nature of this one-off opportunity, and A’s general disposition to comply with the law, A would not have committed any crime if the officers from internal affairs had not trapped him. The reason is that he would never have been afforded the opportunity to commit the only sort of opportunistic crime that he is capable of committing. Yet given that he believed that the opportunity had arisen, he formed the intention to commit the crime. Arguably, the mere possession of an intention—in a context of police provision of opportunity—is not sufficient to justify the integrity test. The reason is not that A is not culpable; clearly A is guilty of an act of theft. Rather, the reason is that integrity testing under these conditions involves the creation of crime, rather than the detecting or preventing of crime that would have existed independently of the integrity test.

Let us take another look at our scenario, but this time let us assume that, unbeknown to the police, Officer A has a disposition to commit opportunistic acts of theft of large amounts of money, if they are left lying around and A believes he will escape detection. But let us further assume that there are no such opportunities. While A hopes for such opportunities, and tells his friends he is waiting for such opportunities, none have been or are ever likely to be forthcoming. As it happens, a one-off opportunity does come, and A is trapped. As before, the objective test does not rule out this kind of trap. Moreover, the subjective test does not rule out this kind of trap either; for A has a disposition to engage in opportunistic theft of large amounts of money.

Notwithstanding the existence of A’s disposition to engage in opportunistic theft of large amounts of money, it still remains the case that A would not have committed any crime if the internal affairs officers had not trapped him. The reason is that he would never have been afforded the opportunity to commit the only sort of opportunistic crime that he is disposed to commit. Accordingly, it might be suggested that the possession of a disposition and an intention—in a context of police provision of opportunity—is not sufficient to justify integrity testing. The reason is that integrity testing under these conditions involves the creation of crime, rather than the detecting or preventing of crime that would have existed independently of the integrity test. On the other hand, presumably Officer A is unusual in being a desk-bound police officer who is not afforded opportunities to engage in opportunistic acts of corruption; many, if not most, police are in fact afforded such opportunities. So arguably if a police officer has a disposition to engage in opportunistic theft then he will engage in it, since opportunities will arise sooner or later.

Let me conclude this chapter by attempting to detail the general conditions under which targeted integrity testing of police is morally permissible. In so doing, I will try to accommodate the various objections made above to entrapment, and to the subjective and objective tests.

First, there are a number of such general conditions, such as the condition that the method of integrity testing is the only feasible method available to law enforcement agencies in relation to a certain type of offence, and that the offence type is a serious one.

Second, the integrity test should be the targeted testing of an officer (or group of officers) who is/are reasonably suspected of engaging in crimes of the relevant kind.

Third, the suspect is ordinarily presented with, or typically creates, the kind of opportunity that they are to be afforded in the integrity test scenario. This condition in large part rules out police creation of crime.

Fourth, the inducement offered to the suspect is: (a) of a kind that is typically available to the suspect, and; (b) such that an ordinary police officer would reasonably be expected to resist it. This condition rules out excessive inducements, and therefore one way in which crime might be created by the police.


Harfield, Clive, and Karen Harfield. 2012. Covert investigation, 3rd ed. Oxford: Oxford University Press.

Kleinig, John. 1996. Ethics of policing. New York: Cambridge University Press.

Marx, Gary T. 1988. Undercover: Police surveillance in America, 1988. Berkeley L.A.: University of California Press.

Marx, Gary T. 1992. Under-the-covers undercover investigations: Some reflections on the state’s use of sex and deception in law enforcement. Criminal Justice Ethics 11(3): 13-24.

Miller, Seumas, and John Blackler. 2005. Ethical issues in policing. Aldershot: Ashgate.

Miller, Seumas, and Ian Gordon. 2014. Investigative ethics: Ethics for police detectives and other criminal investigators. Oxford: Wiley-Blackwell.

Prenzler, T., and C. Ronken. 2001. Police integrity testing in Australia. Criminal Justice: The International Journal of Policy and Practice 1(3): 319-342.

< Prev   CONTENTS   Source   Next >

Related topics