Concept
Limits on Extended Liability
Scope of the Conspiratorial Relationship
Limits on Extended Liability
The sections above fix the reach of a single defendant's conspiracy: the crimes the defendant purposely sought to promote, the partners the defendant agreed with, and the partners' partners that § 5.03(2) adds on proof of knowledge.1 All of that concerns the offense of conspiracy itself. A different question arises when a member of the conspiracy goes on to commit a substantive offense, such as the beating of a watchman during a planned burglary, and the prosecution asks whether every other member answers for it too. Subsection (2) widens the circle of people the defendant conspired with, but it adds nothing to the list of substantive offenses for which the defendant is responsible. The first sub-section below shows how the Code limits liability for a co-conspirator's crimes, and the second shows that trying related conspiracies together leaves both liability and evidence where they were.
Rejecting Pinkerton Liability
The rule the Code rejects comes from Pinkerton v. United States, in which the Supreme Court in 1946 applied agency principles to conspiracy.2 The defendants were two brothers, Daniel and Walter Pinkerton, convicted of conspiracy to violate the Internal Revenue Code and of several substantive violations of it. Walter alone committed the substantive offenses. The record contained no proof that Daniel took part in them, aided them, or knew of them, and Daniel was in prison when some were committed, yet the Court upheld his convictions on those counts.3 The Pinkerton rule that grew out of the case holds a conspirator liable for a crime committed by another member of the conspiracy, even if that crime was not part of the original agreement, so long as the crime was in furtherance of the conspiracy, within its scope, and a reasonably foreseeable consequence of the agreement.4
The reasoning behind the rule treats each conspirator as an agent of the others, so that whatever one does within the venture is attributed to all. Attribution of that kind requires no showing that the defendant wanted the further crime to happen, because foreseeability does the work that purpose does under the Code. Justice Rutledge's dissent attacked the rule at its root. Congress had separately defined completed substantive offenses, aiding and abetting, and conspiracy, and he reasoned that using a conspiracy conviction to supply the first without proof of the second either convicts one man for another's crime or punishes the same man twice for the same offense.5 He also objected that principles of vicarious liability drawn from commercial law and tort law are out of place in criminal responsibility.6
The Code answers that objection directly. It imposes liability on conspirators for the substantive crimes of their co-conspirators only when the strict conditions for accomplice liability are met.7 Those conditions center on the defendant's own purpose, because an accomplice acts with the purpose of promoting or facilitating the commission of the particular offense charged, whether by soliciting it or by aiding in its planning or commission.8 Two questions therefore arise whenever a conspirator commits a further crime, and they have different answers. Whether the defendant conspired is settled by § 5.03(1) and (2), and it yields a conspiracy conviction. Whether the defendant answers for the further crime itself is settled by accomplice liability, and a purpose to promote a burglary supplies no purpose to promote an assault that the defendant never requested and never helped to carry out.
The drafters of the Code explained the choice with People v. Luciano, in which Luciano and others were convicted of sixty-two counts of compulsory prostitution, each count resting on a specific instance of placing someone in a house of prostitution or taking money from the practice, all proved as acts done under a combination to control commercialized vice in New York City.9 Liability was proper for the defendants who directed and controlled the combination, because they solicited and aided numberless specific crimes and so were accomplices under ordinary rules. The result looks different for the prostitutes and runners at the bottom. They were parties to the conspiracy and answerable for their own crimes and for those they aided, but none exerted substantial influence on the hundred others pursuing private ends within the shelter of the combination.10 The mechanism is easy to state. In a large venture the number of offenses grows with the number of participants and the passage of time, while a peripheral member's ability to influence any of them does not grow at all, so liability tied to the conspiracy alone rises with the size of the enterprise and falls out of step with the defendant's own contribution. The drafters warned that the law would "lose all sense of just proportion" if each member were held accountable for thousands of additional offenses of which he was unaware and which he did not influence.11
The Code's position is a minority choice rather than a settled national rule. The federal courts and some states continue to permit vicarious liability for co-conspirators' substantive offenses, while New York and other states, Massachusetts among them, have rejected it.12 The stakes of the dispute appear in United States v. Mothersill, in which defendants convicted of racketeering or drug conspiracy were also found guilty, under the Pinkerton co-conspirator theory, of the death of a state trooper killed by a pipe bomb hidden in a package in the trunk of a rental car.13 State courts keep reconsidering the question. Connecticut reaffirmed the rule on the view that a defendant who played a necessary part in setting in motion a discrete course of criminal conduct cannot fairly complain of liability for its natural and probable results, which the defendant should have foreseen. Washington rejected the rule as inconsistent with the mens rea requirements of its accomplice liability provisions. Nevada split the difference, overturning convictions for a co-conspirator's specific-intent offenses while sustaining those for general-intent offenses, because Pinkerton liability for specific-intent crimes would let the state sidestep the statutory mens rea.14 Even where the rule applies it is not retroactive, so a person who joins an ongoing conspiracy is not liable under it for substantive crimes committed before joining.15
A related doctrine, the common-design rule, attributes to each party to a joint plan the acts that any of them commits in furtherance of that plan. In the Illinois decision in Kessler, three defendants had jointly planned a burglary of a tap, and the burglary produced an attempted murder of the tap owner and of a state trooper who responded; the court held each participant legally accountable for the others' conduct in connection with the planned burglary.16 Because the rule requires a common design or agreement, it works as the functional equivalent of co-conspirator liability, and it confines liability for secondary crimes to those who agreed to the primary crime.17 The Code asks for more than either rule does. An agreement to commit one crime, like the foresight of another, falls short of the purpose to promote the particular offense that accomplice liability requires, which is why a defendant who is a conspirator and nothing more answers for the conspiracy and not for the offense a co-conspirator carried out.
Step three is where answers go wrong, because foreseeability is not purpose. A conspirator who could have predicted the assault has still not solicited or aided it. The conspiracy conviction stands in every version of the facts above, and what changes from one version to the next is whether a second conviction, for the substantive offense, rests on a separate basis.
Joinder Does Not Enlarge Liability
The second limit concerns procedure. Subsection (4)(a) of § 5.03 allows two or more persons charged with conspiracy to be prosecuted jointly in either of two situations: they are charged with conspiring with one another, or the conspiracies alleged, whether they have the same or different parties, are so related that they constitute "different aspects of a scheme of organized criminal conduct."19
Joint prosecution of related conspiracies serves the ordinary reasons for trying connected cases together. The same scheme is proved once, before one jury, instead of repeatedly through overlapping witnesses, and the findings about the scheme do not diverge from trial to trial. The second situation in the provision matters most in large operations, because it lets the prosecution try together conspiracies that have different parties when they are different aspects of one scheme. In an operation of the kind involved in Bruno, the smugglers imported narcotics and each group, smugglers, distributors, and retailers, possessed and sold them, and the Code's approach asks, as to each defendant, whether and with whom that defendant conspired to commit each of those crimes.20 Joinder lets the groups appear in one courtroom, and their separate agreements remain separate.
The danger in joinder is that a jury hearing the whole scheme will treat every defendant as responsible for all of it. A prosecutor has an obvious reason to bundle many defendants before one jury, because scheme-wide evidence then surrounds each of them, and if joinder also changed what each defendant answers for, it would reach through procedure the result the Code refuses to reach through substantive law. Paragraph (b) therefore adds two protections that apply in any joint prosecution. First, no defendant may be charged with a conspiracy in any county other than one in which the defendant entered into the conspiracy or in which an overt act pursuant to it was done by the defendant or by a person with whom the defendant conspired.21 Second, neither the liability of any defendant nor the admissibility of evidence against that defendant is enlarged by the joint prosecution.22 The proportion concern behind the rejection of Pinkerton explains this second clause as well: a defendant's exposure follows the defendant's own agreement and purpose, and sitting beside others who agreed to more changes neither.
The trap is to treat joinder as if it merged the defendants' agreements: joinder enlarges neither liability nor admissible evidence. A joint trial answers where and with whom a defendant is tried. Whether the defendant conspired, and to commit which crimes, is still decided by subsections (1) through (3) on that defendant's own purpose and agreement.
Rejecting Pinkerton Liability
The scope rules in § 5.03 decide with whom, and to commit which crimes, a defendant is guilty of conspiracy. They leave open a further question that arises once one member of the group commits a new offense: whether every other member answers for that offense as well. The majority common-law answer is the Pinkerton rule. It holds a conspirator liable for a crime committed by another member of the conspiracy, even one outside the original agreement, if the crime was in furtherance of the conspiracy, within its scope, and a reasonably foreseeable consequence of the agreement.24 The rule treats partners in crime as agents of one another, so whatever any of them does to carry the venture forward is attributed to all. In a large operation its reach is enormous. A dealer selling a baggie on a street corner is linked by the conspiracy to hundreds of participants whose crimes run from growing poppies to smuggling, bribery, and money laundering, and under the rule the dealer may be punished for each of those crimes as though personally committed.25 The Code rejects this approach outright.26
What the Code puts in its place is the ordinary law of accomplices. A conspirator answers for a co-conspirator's substantive crime only when the strict conditions for accomplice liability are met.27 Those conditions center on the defendant's own purpose, because an accomplice acts with the purpose of promoting or facilitating the commission of the offense charged, either by soliciting it or by aiding in its planning or commission.28 The two routes to liability therefore ask different questions. Conspiracy under § 5.03(1) asks whether the defendant, with the purpose of promoting a crime, agreed to commit it or to aid in its commission.29 Accomplice liability asks whether the defendant had that purpose toward the particular offense the partner carried out and did something to advance it. An agreement to commit one crime supplies the first showing for that crime only, and the foresight that a partner might commit another supplies neither. Pinkerton lets foreseeability do the work of purpose, and the Code withholds that substitution because a person can predict a crime without wanting it to happen.
The drafters of the Code drew their line with People v. Luciano in view. There Luciano and others were convicted of sixty-two counts of compulsory prostitution, each count resting on a specific instance of placing a woman in a house of prostitution or taking money from the practice, all proved as acts done under a combination to control commercialized vice in New York City.30 Liability was proper for the leaders who directed and controlled the combination, because they solicited and aided numberless specific crimes and so were accomplices under ordinary rules. The runners and prostitutes at the bottom stand differently. They are parties to the conspiracy and answer for the crimes they committed and for those they actually aided, but none exerted substantial influence on the hundred others pursuing private ends within the shelter of the combination.31 Leaders and runners belong to the same conspiracy, so membership cannot explain the difference. What separates them is whether each solicited or aided the particular offenses.
The reason for drawing the line there is a mismatch of scale. In a large venture the number of offenses grows with the number of participants and the length of time the venture runs, while a peripheral member's ability to influence any of those offenses stays flat. Liability that follows the conspiracy alone therefore climbs with the size of the enterprise and drifts away from what the defendant did, knew, or could affect. The drafters warned that the law would "lose all sense of just proportion" if every member were held accountable for thousands of additional offenses the member never knew of and never influenced.32 The same instinct drove Justice Rutledge's dissent in Pinkerton v. United States, where Daniel Pinkerton was convicted of substantive offenses his brother Walter committed, some while Daniel sat in prison, on proof of the agreement alone.33 Rutledge reasoned that using a conspiracy to supply the substantive counts either convicts one man for another's crime or punishes the same man twice, and he insisted that guilt for serious offenses should remain personal rather than vicarious.34
The Code's position is a minority choice, and the doctrine remains unsettled. The federal courts and some states continue to impose vicarious liability for co-conspirators' substantive offenses, while New York, Massachusetts, and other states reject it.35 The federal approach produced the convictions in United States v. Mothersill, where defendants convicted of racketeering or drug conspiracy were also found guilty under the Pinkerton theory of the death of a state trooper killed by a pipe bomb hidden in a package in the trunk of a rental car.36 State supreme courts keep revisiting the choice. Connecticut reaffirmed the rule, Washington rejected it as inconsistent with the mens rea requirements of its accomplice-liability statutes, and Nevada overturned convictions for a co-conspirator's specific-intent offenses while sustaining those for general-intent offenses, reasoning that Pinkerton liability for specific-intent crimes would let the state sidestep the statutory mens rea.37 A problem that places the defendant in a Code jurisdiction therefore forecloses Pinkerton liability, while one set in a federal court or a Pinkerton state does not.
A third approach reaches secondary crimes through agreement without using the label of conspiracy. In the Illinois decision in Kessler, three defendants jointly planned a burglary of a tap, and the burglary produced an attempted murder of the tap owner and of a state trooper who answered the report. The court held that each participant was legally accountable for the others' conduct in connection with the planned burglary.38 Because that rule requires a common design or agreement, it is the functional equivalent of co-conspirator liability, confining liability for secondary crimes to those who agreed to the primary crime.39 The Ohio instruction described in Lockett v. Ohio worked similarly, telling jurors that if a conspired robbery and the manner of its accomplishment were reasonably likely to produce death, each plotter was as guilty as the principal offender, as an aider and abettor of the homicide.40 Both approaches treat an agreement to commit one crime, combined with the likelihood of another, as enough. The Code requires more, namely purpose directed at the particular offense.
Students most often lose this issue by letting the conspiracy conviction carry the substantive count along with it. A conspirator who could have predicted the partner's crime has not thereby solicited or aided it, and agreement to one crime is not purpose toward another. The conspiracy conviction stands in every version of the facts above. What changes from one version to the next is whether a separate conviction for the substantive offense rests on its own basis.
Joinder Does Not Enlarge Liability
Subsection (4) of § 5.03 governs what happens when related conspiracies are tried in one courtroom. It authorizes a joint prosecution in two situations: when two or more persons are charged with conspiring with one another, and when the alleged conspiracies are so related that they are different aspects of a scheme of organized criminal conduct.42 The second ground is the broader one, because it allows a single trial of conspiracies whose members differ. The smugglers, distributors, and retailers in a narcotics chain, each group agreeing among themselves without all agreeing with one another, can appear together as parts of one enterprise. The justification is practical: a jury that hears the whole scheme understands what each part contributed, and the system hears the common evidence once instead of several times.
The feature that makes joinder efficient also makes it hazardous. Charging a conspiracy can link dozens of co-conspirators and hundreds of individual crimes in one trial, and the result can be a long and unwieldy proceeding in which the jury struggles to sort out which evidence bears on which defendant, with minor participants seated beside kingpins.43 Evidence rules sharpen the problem, since many jurisdictions admit a conspirator's statements made during and in furtherance of the conspiracy against fellow members.44 A prosecutor therefore has an obvious incentive to bundle defendants, because scheme-wide evidence then surrounds each of them and the strength of the case against the worst actors can color the jury's view of the least culpable. If joining defendants also changed what each of them answers for, procedure would reach the result that subsections (1) through (3) refuse to reach in substantive law, where each defendant's exposure follows that defendant's own agreement and purpose.
The venue limit ties the place of trial to the defendant's own conduct or that of the defendant's partners. A conspiracy consists of an agreement and acts that may be scattered across many counties, so without the limit a prosecutor assembling a multi-defendant case could file wherever any one defendant acted and pull a peripheral member to a place where neither that member nor anyone that member agreed with did anything. The words describing the partner keep the focus individual. An overt act by someone with whom the defendant conspired supports venue, and an act by a partner's partner supports it only where subsection (2) makes the defendant guilty of conspiring with that person.
The no-enlargement clause does the central work in this section, and it follows from the premise that conspiracy liability is individual. A defendant's exposure comes from the defendant's own agreement and purpose as to each crime and each partner, and sitting at a table with people who agreed to more adds neither an agreement nor a purpose. The evidence half of the clause works the same way. A declaration or act of another defendant that could not be used against this defendant at a separate trial cannot be used at the joint one merely because the two are tried together.49 Where a jurisdiction admits a conspirator's statements against fellow members, that exception depends on the declarant and the defendant belonging to the same conspiracy, and a statement from a separate, related conspiracy does not gain admissibility through the shared courtroom.
The third protection supplies a remedy for the risks the first two cannot fully remove. Even with liability and admissibility fixed, a jury that hears months of testimony about other defendants may find it hard to keep each defendant's case apart. A severance gives the requesting defendant a separate trial, and a special verdict has the jury return separate findings on that defendant instead of one general verdict. The residual clause leaves room for instructions and similar steps. The duty is mandatory but conditional: it is triggered by the defendant's request and by the court's judgment that the step is necessary or appropriate to a fair determination of guilt or innocence, so a defendant who simply prefers a separate trial is not guaranteed one.
The trap is to treat a joint trial as though it merged the defendants' agreements into one. Joinder enlarges neither liability nor admissible evidence, so the fact that related conspiracies share a courtroom says nothing about which crimes any one defendant conspired to commit. Joinder decides where and with whom a defendant is tried. Whether the defendant conspired, with whom, and to commit which crimes remains a question for subsections (1) through (3), answered on that defendant's own purpose and agreement.