(concurring in part and dissenting in part). When the immediate interest generated in the wake of these opinions has long since waned, investigators, lawyers, and judges will be analyzing the definition and application of the entrapment defense announced today in order to reach a coherent view of entrapment fully endorsed by a majority of the justices of this Court. I therefore write separately to summarize the force of the opinions taken together and to set forth my position regarding the appropriate approach to the entrapment issue and the reasons for remanding these cases to the trial court.
Initially, it is clear that neither Brown nor Juillet could be found to be entrapped under the objective test. Whether one asks whether a law-abiding person would have given drugs for sex, or whether that person would have provided drugs to third parties because a friend gave him free beer and marijuana, the answer is clearly, "no.”
Second, it is clear that all seven justices favor a new test for entrapment. The objective test for *87entrapment asks whether the police engaged in misconduct of a type that would induce a normally law-abiding person to commit the crime in question. Sorrells v United States, 287 US 435, 459; 53 S Ct 210; 77 L Ed 413 (1932) (Roberts, J., concurring). Under this test, the circumstances of the defendant are irrelevant. Because Justice Brickley’s test considers the circumstances of the actual defendant, and Chief Justice Cavanagh considers the circumstances of the hypothetical defendant (the average drug user), both endorse new tests for entrapment not encompassed in the objective test. Moreover, protestations that the test remains "objective” cannot obscure the fact that neither the prosecution nor the defendants have been given the opportunity to develop a record that anticipates this shift in analysis.
Third, I agree with the force of Justice Brickley’s opinion in People v Jamieson, 436 Mich 61; 461 NW2d 884 (1990), and with Justice Griffin’s partial concurrence and partial dissent that would adopt a subjective test for instigation entrapment. Id., p 98.
Fourth, Chief Justice Cavanagh and I would recognize, to different degrees, a reprehensible-conduct test for entrapment. Finally, in lieu of the Court’s reconsideration of its adherence to the so-called objective test for entrapment and its adoption of the purely subjective test, I would align myself with the version of the entrapment defense set forth today by Justice Brickley.
I
A
Justice Brickley alters the objective test for entrapment by contorting the test so that the trial court may indulge in inferences of susceptibility *88from circumstances of the defendant, such as age, limited education, and dependency. Ante, p 67. The flaw in the approach is immediately apparent when the test, as applied, produces inconsistent results. Although it is objectively more reprehensible to use an addicted prostitute to obtain drugs from a targeted defendant than it is to set up an informant who supplies marijuana to members of a local drug culture, Brown loses and Juillet wins. Since the only distinction between them offered by Justice Brickley is that the government "instigat[ed the] relationship,” in Juillet, ante, p 65, but "only allowed an already existing course of conduct to continue” in Brown, ante, p 64, his result clearly turns on the fact that Brown was ready and willing, that is, predisposed to commit the crime. Justice Brickley’s objective/subjective test nonetheless bars the trial court from considering all relevant evidence, including predisposition,1 that bears on the actual transaction between the police and the defendant and produces results that are neither reliable nor internally consistent.
Chief Justice Cavanagh’s test would even more radically alter the test for entrapment. While barring consideration of the defendant’s individual circumstances and of predisposition, the Chief Justice proposes that we inquire whether the police conduct, objectively speaking, is of a kind that would have caused the average hypothetical drug user to escalate his criminality. He acknowledges that many drug users may escalate on their own; "Some will, some will not,” ante, p 76, n 3. Notwithstanding that the logical force of this observation is to inquire who "will,” or who "will not,” *89Chief Justice Cavanagh would force trial judges to assume a defendant who "will not.” Otherwise stated, at the cost of immunizing the criminal conduct of all those who would escalate, the trial court must assume a circumstance regarding the defendant, in this case, that is, that all drug users, casual, regular, or inveterate, would not escalate. Chief Justice Cavanagh’s test would thus skew the reliability of the entrapment decision by precluding access to any relevant facts regarding the defendant’s role in the transaction.
For the reasons set forth in i(b), I would abandon the objective causation test in favor of the subjective test that permits consideration of all facts and circumstances bearing on whether the government "caused” the conduct in question.
I also would align myself with those justices of the United States Supreme Court and federal and state courts that recognize an objective-misconduct prong of the entrapment defense that bars prosecution in those rare cases in which the government’s conduct is truly reprehensible.
While I do not agree that the entrapment defense should be eliminated, I agree in principle with Justice Griffin’s suggestion that we should "at least move” to the subjective test. Jamieson, supra, p 98. The Chief Justice’s far-ranging analysis will broadly immunize criminal conduct that, in reality, is attributable to the defendants. Justice Brickley’s analysis, though flawed, is the least objectionable of the views that apparently each command three votes. Thus, to provide guidance on the issue, and in lieu of the Court’s adoption of a subjective test for entrapment, I reluctantly align myself with Justice Brickley’s subjective/ objective test for the causation prong of the test.
My colleagues’ contrary contentions notwithstanding, in my view, neither case can be fairly *90reversed or affirmed as a matter of law. Although both opinions profess a logical extension of previous law, in truth each adopts a version of the "objective” test previously unknown in our jurisprudence. Chief Justice Cavanagh does not acknowledge that, under Justice Brickley’s test, both the prosecution and the defendants might now produce evidence that would have been irrelevant under a purely objective test. Justice Brickley does not acknowledge that Brown’s decision not to offer evidence, as well as the prosecution’s failure to introduce evidence (such as the record of the conversations), would have been influenced by the fact that evidence of vulnerability or evidence rebutting the inference of vulnerability was inadmissible under the objective test for entrapment.
Similarly, both opinions state that whatever error was committed below, was error of law in analyzing the evidence of record, correctable in the appellate courts. What each fails to recognize is that, if there was "error in law,” it was error under the newly formulated tests, that a trial judge has not passed on the credibility of evidence offered under these tests and the defendants, and the prosecution, have not been afforded the opportunity to produce evidence for the record relevant thereto or to argue their application to the testimony offered.
In short, because the opinions of the Court today create new tests for entrapment, long after the defendants and the government created records based on the objective test, I would hold that the government and the parties in both cases are entitled to a remand, at which time any party may provide evidence relevant to whether the government’s activities would have induced "a normally law-abiding citizen, in [the defendant’s] circumstances, to commit the crimes with which he’ was *91charged,” ante, p 41, and, additionally in Brown, whether the police conduct was reprehensible in that they manufactured the criminal conduct.
B
Michigan courts have never deviated from articulating a rationale for entrapment founded upon a belief that the doctrine precludes punishment where the government has engaged in conduct that unfairly entices a defendant to commit a criminal act only to prosecute him. We have refused to punish a criminal defendant because "it would seem to be the duty of [an officer of the law] to take such steps as would be likely to prevent the commission of the offense, and tend to the elevation and improvement of the would-be criminal, rather than to his farther debasement.” Saunders v People, 38 Mich 218, 222 (1878) (Marston, J., concurring). The government loses its right to punish when its agents have acted in an extremely reprehensible way.
At the same time, this Court recognizes the necessity for undercover police work and sting tactics to combat virtually undetectable crimes. Thus, the Court has not attempted to decide whether better or more effective investigative techniques might have been used and has avoided a bright-line approach, which can function as authorization per se of all investigative techniques within the bright line as well as immunization of criminal conduct outside the bright line.2 Devising *92a specific test that will preclude punishment where government misconduct unfairly manufactures or creates crime without preventing essential law enforcement activity may not be possible.3 However, I do not agree with Justice Griffin that the courts should abandon the long-asserted authority to restrain abusive investigative conduct. The rhetoric suggesting eradication of constitutional rights as the cost of the war on drugs may, at times, be overblown. That fact should not lead to complacency regarding the lengths to which police frustration and even good intentions may carry a given investigation.
I am persuaded that we should abandon the objective entrapment test of instigation. The conclusion derives some urgency from the inadequacy of the test put forth by Justice Brickley for an "objective test” that focuses on the particular circumstances and vulnerability of the defendant, while barring evidence of predisposition and producing inconsistent results as applied in the instant cases. Likewise, while Chief Justice Cavanagh’s test would postulate "the average hypothetical drug user,” a formulation that ostensibly comports with the objective approach, to state the *93postulate is to expose its fatal flaw. To be sure, there may be situations in which it is possible to answer the inquiry regarding whether the government’s conduct "instigate[d] the commission of a crime by one not ready and willing to commit it,” Stewart, J., dissenting, United States v Russell, 411 US 423, 445; 93 S Ct 1637; 36 L Ed 2d 366 (1973), but the hypothetical casual drug user assumption will not yield a reliable answer regarding whether government conduct caused the average drug user to escalate his criminality "where such escalation would probably not otherwise have taken place.” Ante, p 78. In point of reality, we know nothing regarding whether a defendant was a casual drug user, a regular drug user, or an inveterate abuser. The irreducible reality is that, no matter how it is labeled, an instigation test examines the defendant’s characteristics. The inquiry is subjective and necessarily so. The question is whether the defendant would have done the act without government encouragement. Thus, most importantly, I would adopt the subjective test for instigation because failure to define entrapment by reference to both subjective and objective factors, distorts any meaningful evaluation of the reality of what caused the transaction between government and accused.
However, because government conduct may be reprehensible, not only because it manufactures crime, but on the basis of the level of misconduct alone,4 I would embrace a dual view of "entrapment” that would bar prosecution because of truly *94reprehensible police conduct and would bar prosecution for conduct that instigates or manufactures crime. Government misconduct would be evaluated by focusing on police conduct. Government instigation would be evaluated subjectively by taking into account the defendant’s circumstances and the interaction between the government agent and the defendant, and the trial court would determine both issues as a matter of law. Such an approach will accommodate the necessity to protect against overzealous law enforcement that is the heart of the objective approach, Sherman v United States, 356 US 369, 381; 78 S Ct 819; 2 L Ed 2d 848 (1958) (Frankfurter, J., concurring in the result), while preserving the opportunity for the defendants who claim that, although the police conduct was not intolerable in the objective sense, it was intolerable in light of the particular circumstances.
The objective misconduct defense is justified when the government engages in illegal behavior, the government conduct invades personal privacy, the government unfairly manufactures antisocial activity, or "because [entrapping behavior] places in the hands of the executive the power to make criminals.” Seidman, The Supreme Court, entrapment, and our criminal justice dilemma, 1981 Sup Ct R 111, 145.
Government encouragement of an individual to perform a prohibited act interferes with the functions expected of the act requirement in a rule-oriented system of criminal law. Carlson, The act requirement and the foundations of the entrapment defense, 73 Va LR 1011, 1056 (1987). To *95merit punishment, criminal conduct requires both an actus reus, a voluntary act or omission which causes social harm, and a mens rea, the mental state provided in the definition of the offense. See, generally, Dressier, Understanding Criminal Law, pp 63-115. The act requirement embodies critical limitations on the imposition of criminal sanctions. It guarantees that the accused has done something to merit punishment,5 preserves the sphere of personal autonomy in which an individual can think and act without fear of government intrusions,6 and prevents arbitrary and abusive exercise of government power in the criminal law arena.7 Carlson, supra, pp 1053-1056. Because the use of government encouragement thwarts these protections of the act requirement, the entrapment defense prevents criminal prosecutions when government authority has been undermined by reprehensible government conduct._
*96The subjective component of the entrapment defense recognizes that trial courts cannot realistically determine whether the governmental activity in question has manufactured the crime without considering the circumstances surrounding the transaction.
As Jamieson explicitly recognized, a judge cannot reliably answer the question regarding whether the conduct in question was caused by the police by postulating only a defendant’s vulnerability or the hypothetical defendant. The normally law-abiding or average person probably cannot be induced or tempted into criminal conduct absent duress or coercion.8
An objective formulation of the causation prong may be unreliably underinclusive because it eliminates susceptible persons who succumb to weaker pressure or inducements than the normal hypothetical person. Yet, an inducement which seems "fair in the abstract may be unfair in a particular case, for reasons that are unknown to the agent and therefore do not affect the propriety of his conduct.” Park, The entrapment controversy, 60 Minn LR 163, 220 (1976). Likewise, the test may be unreliably overinclusive because it requires a trial court to determine the instigation question *97without considering probative evidence bearing on the issue.9 This Court’s historical concern with misconduct is addressed by recognition of a variation of the due process defense that will bar prosecution when the government has engaged in truly reprehensible conduct.10 This approach will preserve the Court’s historic authority to curb abusive police conduct while permitting a realistic examination of the question regarding whether government has instigated the criminal conduct. Saunders, supra, p 222.
Under an objective-misconduct prong, the trial court focuses on the government conduct. Government inducements should not make the crime unusually attractive nor, as Jamieson explained, should "discretionary investigative enforcement measures extend beyond a tolerable level when by design the government uses continued pressure, appeals to friendship or sympathy, threats of arrest, an informant’s vulnerability, sexual favors, or procedures which escalate criminal culpability.” Jamieson, supra, p 89. (Emphasis in the original.) The inquiry is whether the government’s conduct *98falls below an acceptable standard for the fair and honorable administration of justice.
The subjective analysis inquires into whether the law enforcement conduct instigated or created the crime, an inquiry which involves an analysis of all evidence bearing on the question, including predisposition.11 Because the causation question cannot be answered without reference to the defendant on trial, this inquiry has been, in fact, part of the objective test. Sherman, supra, p 384 (Frankfurter, J., dissenting).12 See also People v Turner, 390 Mich 7, 23; 210 NW2d 336 (1973).
Since the government-instigation or -causation prong considers whether a given act accurately gauges the threat that the defendant poses to society absent police encouragement, see Carlson, supra, p 1093, n 254, the court must examine the transactions leading up to the offense, the inter*99action between the agent and the defendant, the type and size of inducements, the nature of the criminal conduct at issue, the extent to which the investigation targets a particular individual, and the extent to which procedures used escalated criminal culpability of the defendant. The defendant’s predisposition is a relevant factor under this prong since it bears upon whether he would have posed a threat to society independent of government inducements.13 It is not dispositive, but the instigation issue cannot be accurately evaluated14 if weaknesses must be assumed, if prior unrelated criminal activity must be disregarded, and if motive to commit present activity may not be inferred from past similar activity.15 Just as improper police conduct weakens the authority of *100government to punish, trial court findings based on jerry-built constructs weaken the factual reliability of the trial courts’ findings and, hence, public confidence in the result.
II
These cases confront the Court with highly disparate factual scenarios. In Brown, the Court is presented with an apparently targeted investigation into a well-known public figure in which the government informant, a prostitute and drug addict, gained access to Brown’s apartment to obtain evidence of the possession and delivery of drugs through her longstanding relationship with Brown. In Juillet, the Court is presented with an unfocused broad-ranging undercover investigation into drug use among young persons in Cheboygan County into which the defendant stumbled after meeting the informant on the street. In each of these cases, the defendants contend that government conduct was sufficiently reprehensible to warrant an entrapment finding.
A. PEOPLE v BROWN
Having set forth the analysis of entrapment I would adopt, and having agreed that in lieu thereof I would adopt Justice Brickley’s objective/ subjective test, I must disagree with the conclusion which would affirm the lower court findings that Brown was not entrapped. In my view, the record below supports neither Justice Brickley’s conclusion that Brown was not entrapped, nor Chief *101Justice Cavanagh’s conclusion that the government escalated the criminal activity or engaged in reprehensible conduct. I would reverse the decision of the Court of Appeals and remand the case for further hearing and findings regarding whether entrapment occurred under the revised standards for the entrapment.
Brown sought to have the trial court consider the significance of evidence regarding government misconduct without consideration of his individual circumstances. Both the trial court and Court of Appeals shifted focus from government misconduct to whether the government instigated the criminal act of which Brown was convicted. By analyzing the case under the government-instigation prong, despite the lack of evidence presented, the lower courts overlooked consideration of the government’s mistreatment of the informant, Roberts.
The trial court stated the test for entrapment as whether the record supports "a conclusion that the police or their agent manufactured Defendant’s crimes by conduct likely, when objectively considered, to induce or instigate the commission of a crime by a person not ready and willing to commit it . . . .” Yet, when applying the test to the facts of this case, the trial court focused on Brown’s predisposition, that is, whether the police changed Brown’s previous course of conduct. This was error under the objective test, and it is error under the lead opinion’s merging of objective and subjective considerations.
Justice Brickley rests affirmance of the case on the apparent conclusion that Mr. Brown was predisposed, ante, p 62, without consideration of the subjective vulnerability that he is willing to indulge regarding defendant Juillet. Brown is denied relief because "the solicitation was no different than the many offers made in routine purchases *102and sales by police informants,” id., while Juillet’s conviction is barred despite the fact that he committed the oifense with which he was charged within hours of meeting the informant. It is true that Brown did not testify that he was addicted to drugs, id., p 63. Neither did Juillet. Nor is there evidence in the entrapment hearing in Juillet of dependency or appeals to sympathy.16
Finally, Justice Brickley’s analysis does not explain why police encouragement that leads to an informant’s decision to "resume contact with Brown,” id., p 64,17 is permissible use of friendship, while use of "police-manufactured friendship” to investigate the drug subculture in Juillet is impermissible.18 Id., p 65.
The due process defense accepted in United States v Russell, supra, pp 431-432, and acknowledged in the federal courts, is broad enough to consider treatment of the informant and is distinguishable only by degree from the objective standard for entrapment. Note, Entrapment defense in New Jersey, 21 Rutgers L J 419, 435, n 136 (1990). On remand, the court should consider: whether the police themselves engaged in criminal con*103duct;19 whether the police made appeals to past friendships, temptation of financial gain, or repeated solicitations; and whether the police had a legitimate motive for conducting the investigation. People v Isaacson, 44 NY2d 511, 521; 406 NYS2d 714; 378 NE2d 78 (1978).
As Justice Powell observed in Hampton v United States, 425 US 484, 495, n 7; 96 S Ct 1646; 48 L Ed 2d 113 (1976), it will be the rare case in which the misconduct defense will succeed. Vigilance and increased sophistication are essential to effective law enforcement, and the litany of woe in the wake of our drug dominated culture admits of no less. However, consideration of the conduct of law enforcement toward an informant is a factor which may be evaluated in determining whether the government’s conduct falls below an acceptable standard for the honorable administration of justice.20
The court on remand should address under the *104government-instigation or -causation prong all relevant considerations bearing on the instigation question, including, but not limited to, who initiated the sexual relationship, who first introduced drugs as a method of payment, and the extent to which, if at all, the contents of the tapes of Roberts’ conversations bear on the instigation or misconduct issue.21 The. lower courts would also consider the most troubling aspect of the case under the misconduct analysis — the government’s treatment of its informant, Roberts, a known prostitute, heroin addict, and user of other illicit drugs.22 The trial court should also address whether the government agreed that Roberts could inject herself with cocaine and whether it was necessary to obtain the evidence. Under the tests adopted today, the trial court must determine whether, on the basis of all the facts and circumstances, the government instigated the defendant’s conduct and whether the government’s conduct manifested an exploitation of Roberts that is inconsistent with the proper use of governmental authority.23
B. PEOPLE v JUILLET
I also disagree with the finding that the defen*105dant Juillet was entrapped.24 However, although five justices of this Court are prepared to bar the prosecution on the basis of entrapment, I observe that this reversal bodes far more serious consequences for effective enforcement of the law than the result in People v Brown. Although symmetry requires remand of this case for further hearing and findings regarding whether entrapment occurred under the revised standard for entrapment,25 there is on this record no evidence suggesting government conduct that was offensive in the objective sense.
In addition to the delivery of marijuana to Trooper Patrick, Juillet was charged with delivery of marijuana for his conduct in a transaction occurring on March 2, 1982. On that occasion, he met Bleser and another undercover agent and arranged a meeting with a dealer who had marijuana and lsd for sale.
Despite extensive testimony regarding the police undercover operation and Bleser’s conduct, there is no evidence that the size or type of inducements used by Bleser resulted in an opportunity to commit crime that was uncommon or excessive. Nor was there unambiguous testimony that Juillet acted out of a motive, such as sympathy or fear, other than ordinary criminal intent. Although Juillet testified that he considered Bleser his *106friend, the record reveals an association between the two that had begun less than a month before the transaction giving rise to this action. Juillet also remarked that Bleser frequently expressed interest in purchasing drugs for himself or in assisting friends to purchase drugs. But there is no suggestion that Bleser urged Juillet to help because of their friendship. Nor is there any testimony suggesting that Juillet thought Bleser was addicted to drugs or that he or his friends were in need of drugs to avoid withdrawal. Finally, while Justice Brickley seems to believe that the defendant was convicted of being a dealer in drugs, he was not. Juillet, like Brown, was convicted of delivery of marijuana, an offense that does not require consideration . as an element. Juillet pleaded guilty to delivery of marijuana, the precise activity he engaged in on the very first day he met Bleser, and with little or no inducement. Objectively viewed, it appears that the police simply set a widespread net into which Juillet unsuspectingly fell.26 See Folsom v State, 734 P2d 1015, 1017 (Alas App, 1987).
Allegations were made that Bleser repeatedly violated various laws, including those against use and delivery of drugs. The facts suggest that Juillet engaged in the delivery of marijuana on a small scale and not for profit. They do not suggest that Juillet never delivered drugs to other persons in exchange for money or goods. Apparently, Juillet sold drugs for cost rather than to make a profit, but the statute does not require a profit for a violation. MCL 333.7401(1); MSA 14.15(7401)(1). While Bleser’s activities may have been unaccept*107able in the overall scheme, they do not appear to have been relevant to Juillet’s conduct in the specific charge.27
To reemphasize my conclusion that the result in Brown can be neither reversal nor affirmance on the record as it stands, I observe again that Justice Brickley finds that Brown was predisposed and Juillet was not. To the extent that these conclusions rest on circumstances of the defendant, these are mixed questions of fact and law that the prosecution has been given no opportunity to contest in Juillet, and that Brown had no opportunity to explore under the objective test. Indeed, Justice Brickley’s conclusion that, since there is no evidence that Juillet was a drug dealer, the only reason for his "delivery of drugs” was Bleser’s incessant requests, ante, p 68, is classical question begging. Under the objective test, extant in 1982, Juillet’s predisposition, like that of Mr. Brown, was irrelevant. Thus, whether either was a drug deliverer was irrelevant and inadmissible in the entrapment hearing. To conclude from the absence of an inadmissible fact that the government must have manufactured the crime in Juillet but did not manufacture the crime in Brown again evidences that the newly created objective/subjective test will not yield a reliable result, and that it will allow questionable conduct such as that in Brown, while prohibiting routine undercover activities such as that in Juillet.28_
*108Because the opinions of the Court today have adopted new tests for the definition of entrapment, I would reverse the Court of Appeals and remand the case to the trial court to afford the parties the opportunity to present any additional evidence relevant to the revised standards.29
III
I would reject, under the instigation test, the approach that hypothesizes a defendant without a history as well as the fantasy that creates the "average, hypothetical drug user.” Cavanagh, C.J., ante, p 78. The approach that accords with reality and is, therefore, most likely to produce a reliable result is to analyze the facts under the government-instigation prong to determine whether the criminal conduct accurately measures the threat to society from the individual defendant, or whether the circumstances suggest that the defendant, if left alone, would not have engaged in the criminal conduct. I would recognize a government-misconduct defense that bars prosecution where the evidence has been obtained by reprehensible conduct, objectively evaluated.
I have opted to align myself with the instigation analysis of the lead opinion because to leave the Court in its divided posture is to do, perhaps, greater damage to the Court and the jurisprudence than I fear' from either of the new "objective” tests. I cannot agree, however, that either the *109government or the defendants should be punished or should benefit from the fortuity of the various tests announced today. I would, therefore, reverse the decisions of the Court of Appeals and remand Brown and Juillet to the trial court for further proceedings.30 I recognize that both the defendants and the government have a legitimate interest in the long delayed resolution of these cases. In my view, however, giving each an opportunity to formulate their positions around the opinions issued today better serves both their respective interests and that of the jurisprudence, than to say one side wins or loses because of a failure to perceive the evolution of the "objective test” of Jamieson or the penumbras its shadow casts today.
By contrast, I read Jamieson as permitting consideration of all the circumstances surrounding the transaction and as rejecting the prosecution’s argument that the entrapment issue should be a jury question.
A formulation limiting the probative force of the defendant’s circumstances to his weaknesses might invite the innovative defendant to employ runners who are young and have "limited education,” ante, p 67, a result which Justice Matthews of the Alaska Supreme Court illustrated as permitting "drug sellers to insulate themselves from conviction by the device of requiring all their customers to grovel briefly before a sale is made.” Pascu v State, 577 P2d 1064, 1069 (Alas, 1978) (Matthews, J., concurring).
Judicial line drawing is a task which the courts often encounter, but, in examining the interplay between criminal conduct and government investigatory action, the search for general principles to guide the court is especially difficult. Professor Kent Greenawalt observed that when confronted with serious moral choices "we test our intuitive reactions to particular situations against our accepted principles” until both may give, and "we arrive at what John Rawls calls a 'reflective equilibrium,’ in which our sense of right for particular issues matches our principles.” Greenawalt, The enduring significance of neutral principles, 78 Colum L R 982, 997 (1978). It is that process which has sometimes caused this Court difficulty in the past and in which we are engaged in these cases. Nevertheless, the common law is built upon the gradual development of precedent as courts mesh guiding principles with shifting fact patterns which arise as circumstances change. Given the Court’s duty to prevent prosecutions on the basis of reprehensible police conduct, it is necessary to examine each new set of facts in light of principles articulated by this Court.
In observing that there may be circumstances in which conduct not directed at a predisposed defendant might bar the government from invoking the judicial process, Judge Henry Friendly observed, "there 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 .... '¡investigation’ involving participation in activities that result[s] in injury to the rights of its citizens is a course that *94courts should be extremely reluctant to sanction.” United States v Archer, 486 F2d 670, 676-677 (CA 2, 1973). It is similarly unthinkable that any one of us would opine that Brown’s conviction could stand had the government tortured or threatened an unwilling Roberts to gain evidence against him.
Whether adhering to a retributive or utilitarian justification for punishment, the act requirement ensures that punishment is justifiable because of the social harm committed by performing the act itself or the social harm resulting from a violation of the social contract or as an indicator of the potential dangerousness of the criminal actor. See Carlson, supra, pp 1059-1082. It has been suggested that cases like that involving John De Lorean in which the jury found entrapment, despite evidence of predisposition, rest on the factfinder’s sense that the moral authority of government to punish has been undermined by the methods employed. Comment, Entrapment, De Lorean and the undercover operation: A constitutional connection, 18 J Marshall L R 365 (1985).
Government employs its power to intrude into an individual’s decision making rather than maintaining a "neutral position vis-á-vis its citizens and the choices that they make.” Carlson, supra, p 1086. This forces the individual to expend time and energy on "unsolicited and probably unwanted enticement to crime, and he is forced to make a choice that he otherwise might never have faced.” Id., p 1085.
Generally speaking, the government is precluded from imposing criminal sanctions until there has been an actus reus. The act requirement limits the government to punishing conduct which causes some social harm, "provides a reasonably objective basis for assessing who should be punished, and hinders the government’s ability to target its enemies or protect its friends when it uses the criminal sanction.” Carlson, supra, p 1089.
Pasen v State, n 2 supra, p 1067 (replacing the Grossman v State, 457 P2d 226 [Alas, 1969] average-person standard with a focus on the particular conduct of police in the case presented); State v Cripps, 692 P2d 747, 749-750 (Utah, 1984) ("an average person or ordinary citizen is not a former drug addict, will not be begged by a former lover to obtain drugs, does not have any notion of how to reach people who sell drugs, would probably not befriend the sort of stranger who turns out to be an undercover narcotics agent, and could not imagine circumstances short of physical threats that would prompt him to obtain marijuana to sell”). Note, People v Barraza: California’s latest attempt to accommodate an objective theory of entrapment, 68 Cal L R 746, 757 (1980) (characterizing the explanation of entrapment on the basis of normally law-abiding persons as circular and concluding that it is questionable whether the standard can be applied without "emasculating the entrapment defense”).
For example, if the government targets a political foe, the conduct may warrant an entrapment finding, but resolution of the question necessarily requires examination of both the government conduct and the characteristics of and government interaction with the individual defendant. Likewise, judges may intuitively suspect that some of those defendants charged with major drug offenses may have reached the amount that triggers mandatory minimum sentences because the amount, rather than the criminal intent of the defendant, was the target of the investigation, another inquiry which involves a factual evaluation of the characteristics of the defendant and the interaction between the defendant and the government.
Five justices comprising the concurring and dissenting opinions in Hampton v United States, 425 US 484; 96 S Ct 1646; 48 L Ed 2d 113 (1976), expressly recognized the viability of a due process defense in cases involving outrageous government conduct. As developed in the lower federal courts, the defense is generally available to a defendant admittedly predisposed to commit the crime and raises a question of law for the court. United States v Twigg, 588 F2d 373 (CA 3, 1978); Klar, The need for a dual approach to entrapment, 59 Wash U L Q 199 (1981).
One who raises the instigation defense should not be regarded as implicitly arguing that the criminal conduct, rather than being a free and deliberate act, was caused by the inducements of the police. Such an analysis is contrary to basic assumptions of criminal law concerning the free will of the defendant and confuses causal statements based on the idea of instrumentality with causal judgments meant to reflect understanding of the motivation of the individual. See comment, Causation and intention in the entrapment defense, 28 UCLA L R 892-894 (1981). The type of inducements ordinarily at issue when entrapment is raised does not involve the use of an individual as an instrumentality, that is, without his exercise of will or voluntary action. Normally, the government provides inducements which tempt or persuade the defendant to engage in the complained-of conduct but the defendant remains free to act or not. When persons are used as an instrumentality of another, they are normally excused from responsibility for their actions. On the other hand, the law does not normally excuse an actor simply because it understands the forces that motivated him to engage in the proscribed conduct. To hold otherwise could result in a universal excuse. Dressier, Reñections on excusing wrongdoers: Moral theory, new excuses and the modern penal code, 19 Rutgers L J 671, 687 (1988).
See also Hampton v United States, n 10 supra, p 494 (Powell, J., concurring) (under the objective test of the due process approach, differences between the circumstances would be irrelevant despite outrageous government conduct relative to the circumstances, and although predisposition could be proved against persons of widely varying degrees of criminality).
Focusing exclusively on the defendant’s predisposition is unlikely to yield a correct result in some cases. Normally, the criminal act limits punishment in an anticipatory liability situation (such as attempt or conspiracy) to circumstances where the act serves the evidentiary function of corroborating the actor’s criminal purpose. Carlson, supra, pp 1080-1081. When the government has intervened to entice criminal conduct, the evidentiary function of the act requirement has been weakened. If the enticement was unusual or calculated to induce a motive besides ordinary criminal intent, even if the defendant had a long criminal history, it is not always clear that the defendant would have engaged in the conduct without government encouragement. Therefore, even courts adhering to a subjective approach have considered the nature of the inducement and the defendant’s reaction to it as part of the inquiry into entrapment. See, e.g., Sherman v United States, supra, p 373.
While the defendant bears the burden of proving by a preponderance of the evidence that he was entrapped under the objective standard, People v D’Angelo, 401 Mich 167, 180-183; 257 NW2d 655 (1977), federal and state courts applying the subjective test have employed a variety of tests, including placing the burden of production on the defendant and shifting the burden of persuasion to the government. Marcus, The entrapment defense and the procedural issues: Burden of proof, questions of law and fact, inconsistent defenses, 22 Crim L Bull 197 (1986).
The focus on predisposition has been criticized by this and other courts for suggesting that the defendant’s past record can justify police misconduct. See People v Sinclair, 387 Mich 91, 116-120; 194 NW2d 878 (1972); Saunders v People, supra; Sorrells, supra, pp 458-459 (Roberts, J., concurring); note, People v Barraza, n 8 supra. However, while predisposition does not justify police misconduct, the *100government-instigation or -causation entrapment defense posits that the defendant would not have committed the crime without the illegal conduct of the police. The question for the court is, would he? —a question that cannot be answered without a full examination of the defendant and his circumstances.
Both Brown and Juillet apparently had engaged in the use of various drugs in the presence of the government agents. From the initiation of the investigation, both Brown and Juillet supplied drugs to the government agents, Roberts and Bleser, with little apparent hesitation or reluctance. Although Brown and Juillet supplied drugs to the government agents on several occasions, neither was charged with drug dealing nor can be properly characterized as drug dealers.
Although the trial court apparently satisfied itself that there was no governmental misconduct or impermissible motive in targeting Brown, there is no in camera record supporting this conclusion. On remand, the trial court should examine the government’s basis for targeting Brown to insure that no misconduct was involved.
There is no doubt that Brown’s conduct was sordid. He met Roberts as a prostitute, through another prostitute. In the course of the instant investigation, he, Roberts, and another woman used drugs in his apartment. Moreover, while the government may have exploited Roberts’ addiction, so did Mr. Brown. What is sordid conduct on Mr. Brown’s part, may be intolerable conduct on the part of the government.
See People v Taylor, 599 P2d 496 (Utah, 1979) (an undercover agent, an addict who supported her habit by prostitution, had cohabited with the defendant, and resumed the relationship with him to gain evidence for charges of distributing heroin was held to have entrapped him on the basis of her close personal relationship engendering sympathy and pity for her need for drugs).
State courts have upheld convictions despite the use of intimate sexual relationships in cases in which the primary charge is prostitution and the relations are necessary to obtain evidence, but these facts are not before us here. See, e.g., State v Tookes, 67 Hawaii 608; 699 P2d 983 (1985) (the deceitful use of sex by civilian volunteer acting at the behest of police investigating prostitution ring was not outrageous conduct); Anchorage v Flanagan, 649 P2d 957 (Alas App, 1982) (use of sex by an undercover officer in the investigation of prostitution was not outrageous); State v Putnam, 31 Wash App 156; 639 P2d 858 (1982) (use of sex by a civilian working with police in an investigation of prostitution was permissible).
This is not to say that Brown did not instigate the conduct or that the government engaged in such reprehensible conduct that the prosecution must be barred, questions that my colleagues urge me to resolve, albeit to agree with their results. It is to say that both Brown and the government (whose status as an interested party both the lead opinion and dissent overlook) are entitled to an opportunity to formulate the record under the new tests my colleagues adopt.
While it may be difficult to hypothesize a case short of a violation of due process that would violate the misconduct prong where the defendant was predisposed to commit the crime, I would not eliminate that possibility here.
To emphasize the determination that should occur on remand, I observe both my disagreement with Justice Brickley’s conclusion that the government’s treatment of Roberts is irrelevant and the factual conclusion of the Chief Justice regarding the informant’s "vulnerability,” Cavanagh, C.J., ante, p 82.
See People v Isaacson, supra, pp 522-523 (an "overriding police desire for a conviction of any individual” which was obtained by police brutality and "cunning subterfuge employed to enlist the services of an informant,” viewed together, "reveal a brazen and continuing pattern in disregard of fundamental rights” sufficient to preclude prosecution on due process grounds).
I find Justice Brickley’s reversal of this case inconsistent with the objective person, ready and willing to commit the crime, set forth in Jamieson, just as this result is internally inconsistent with the conclusion that Brown was predisposed. Ante, p 62. As noted above, neither Juillet nor Brown were drug dealers. Moreover, to the extent that Juillet’s conduct can credibly be described as simply " sociable’ with another drug user as part of a friendship,” id., p 68, so was Brown’s.
Without prejudging the outcome, I suspect that, in the long run, the jurisprudential significance of today’s decisions will not arise from an unusual situation like that in Brown but, rather, from the conclusion of entrapment of an individual who fell unsuspectingly and rather enthusiastically into a not atypical investigative net.
Bleser struck up an acquaintance with Juillet on the street and, after asking if Juillet was old enough to go to a bar, offered to purchase beer for the two of them. Juillet and Bleser went to Bleser’s apartment where, after drinking some beer, at Bleser’s request, Juillet supplied Bleser with two joints.
This case is unlike Brown, where the government conduct vis-á-vis the informant was directly connected with obtaining the evidence for Brown’s conviction.
Juillet’s strongest argument for an entrapment finding under the instigation prong is that the government took advantage of his drug dependency and limited intelligence to instigate crime that otherwise would not have occurred. However, this argument also lacks adequate record support. Juillet testified that he had never used lsd and had used pcp approximately twice. As the record now stands, Juillet appears only to have been a regular user of marijuana. Nor was *108evidence presented regarding Juillet’s limited intelligence during the entrapment hearing.
The trial court concluded that Bleser did exploit Juillet’s vulnerabilities (by smoking marijuana with him and using his superior intelligence and training) to induce him to sell pop to an undercover agent. The trial court noted that Bleser had smoked marijuana with Juillet and had "causally orchestrated the entire transaction . . . .” We do not reach the propriety of this finding and conclusion since it is not before the Court.
I wish to make clear, in connection with the misconduct question, that I disassociate myself from the unnecessary dicta of the Chief Justice that the entrapment doctrine is rooted in the Due Process Clause of the Michigan Constitution.