We’re here around the clock to help you take the first step toward resolving your legal issue. Reach out anytime for a free initial consultation.
Let us know a convenient time and we’ll get in touch—no waiting on hold.
There is no offence called joint enterprise. No section of any Act creates it, and no separate body of law governs it. The phrase describes ordinary secondary liability: the rule, in section 8 of the Accessories and Abettors Act 1861, that anyone who aids, abets, counsels or procures an offence is tried and punished as a principal offender. What makes joint enterprise feel like a distinct doctrine is how it is used: to prosecute several people together for one act that only one of them committed.
On 18 February 2016 the Supreme Court held in R v Jogee that the courts had been applying that rule wrongly for more than thirty years, and that the previous case law had been wrongly interpreted. This is a plain-English account of what English law requires now, what Jogee changed, what it deliberately did not change, and where a joint enterprise case is actually decided, which is almost never in the courtroom.
It is written for anyone arrested, interviewed or charged over an incident involving other people, for their families, and for professionals advising them. It covers the elements the prosecution must prove, the history of the doctrine, the 2026 reform position, sentencing, and the practical steps that decide these cases.
The criminal law has always allowed more than one defendant to be convicted of a single offence. Someone drives the car, someone holds the door, someone carries the weapon, someone uses it. Section 8 of the Accessories and Abettors Act 1861 puts them on the same footing:
Whosoever shall aid, abet, counsel, or procure the commission of any indictable offence, whether the same be an offence at common law or by virtue of any Act passed or to be passed, shall be liable to be tried, indicted, and punished as a principal offender.
That is the whole of the statutory basis. Everything else is case law. The one who commits the offence is the principal, usually written as D1; anyone liable for assisting or encouraging it is a secondary party, D2. Section 44 of the Magistrates' Courts Act 1980 does the same job for summary offences.
Three things follow from the wording, and all three are routinely misunderstood by people interviewed about a joint enterprise case.
The first is that a secondary party is convicted of the same offence as the principal, not a lesser one. There is no offence of "being an accessory to murder" carrying a lower sentence. An accessory convicted of murder is guilty of murder.
The second is that the prosecution does not have to prove who did what. Where the evidence shows that each defendant present was either the principal or a secondary party, and the elements are made out either way, the jury need not decide which. This is the feature that does the most work in a multi-handed case, and it is why someone who did nothing physical can be tried for a killing.
The third is that joint enterprise is not a legal test. It is a legal term of convenience, commonly referred to as though it were a common law doctrine in its own right, and nothing turns on whether a court calls a case a joint enterprise. The Crown Prosecution Service renamed its own guidance for this reason: it is now "Secondary liability: charging decisions on principals and accessories", last updated on 4 July 2024. The elements are the elements whatever the case is called, and calling it a common law doctrine adds nothing.
Joint enterprise is not new. The principles of accessory liability in the criminal law go back centuries; the 1861 Act consolidated them in the nineteenth century, and its history explains why joint enterprise still provokes such anger. In November 1952, during a break-in at a Croydon warehouse, 16-year-old Christopher Craig fired the fatal shot that killed PC Sidney Miles. His companion, 19-year-old Derek Bentley, was already under arrest when the shot was fired. Bentley was convicted of murder as a joint participant and, because he was an adult and the death penalty still applied, he was hanged in January 1953. Craig, too young to hang, was detained and later released. Bentley's conviction was quashed by the Court of Appeal in July 1998, more than forty years after his execution.
The modern criminal law took its decisive turn in 1984. In Chan Wing-Siu v The Queen [1985] AC 168, the Privy Council held that a participant in a criminal venture who foresaw that an accomplice might commit a more serious crime was guilty of that crime if it was committed. The House of Lords adopted Chan Wing-Siu in English law in R v Powell; R v English [1999] 1 AC 1, and for more than three decades foresight of a real possibility was treated as enough, a lower threshold than intention. That rule became known as parasitic accessory liability. Chan Wing-Siu was a Privy Council decision from Hong Kong, but English law treated Chan Wing-Siu as binding in practice until the Supreme Court and the Privy Council sat together in 2016 to correct it.
Crossfire cases show how far the doctrine stretched. In R v Gnango [2011] UKSC 59, two men exchanged fire in a south London car park, and a passer-by, Magda Pniewska, was killed by the fatal shot, which was fired not by Gnango but by his opponent. The Supreme Court upheld Gnango's murder conviction: although he had not fired the fatal shot, he was held to have committed murder through the shared exchange of cross fire. It was decided before Jogee, and its reasoning on cross fire, where the death results from the other side's bullet, has been debated ever since. Cross fire cases remain among the hardest in the criminal law.
Lumping them together is most of the confusion, because criminal liability arises differently in each.
Joint principals. Two or more people join in committing a single crime and each does enough to commit it. Three people kick a victim; each is one of the joint principals in the assault. No secondary liability arises at all, and the question of who caused which injury does not have to be answered.
One defendant assists or encourages another. D2 drives D1 to the scene, supplies a weapon, keeps watch, or urges D1 on. D1 commits the offence. D2 is liable as a secondary party for that offence, provided the mental element is made out. Most such cases involve assistance of a fairly ordinary kind.
Crime A and crime B. Two or more persons participate together in one crime, crime A, and in the course of committing crime A, the person who goes further, D1, commits a second crime, crime B, which is usually more serious. A fight breaks out, which is crime A, and one participant produces a knife and kills, which is crime B. Whether D2 is guilty of crime B is the question joint enterprise cases are really about, and it is the question Jogee answered.
In practice, joint enterprise most frequently arises in group-related offences such as gang violence, street fights and robberies, where several people are present and one goes further, committing crime B in the course of committing crime A. The first two categories were never controversial. Only the third was, and only because of how the courts had come to approach it.
The elements split in the usual way, into an actus reus and a mens rea, and the mens rea is where contested cases live. The actus reus is conduct capable of assisting or encouraging. The mens rea has three parts: an intention to assist or encourage, knowledge of the essential facts, and, where the offence requires a particular intent, an intention that the principal should act with it.
The conduct element. D2 must have done something that assisted or encouraged D1's offence. Words will do. So will deeds. So, in the right circumstances, will presence, where the evidence shows that being there was itself encouragement, for example because D2 was part of a group whose numbers emboldened the principal. The assistance need not be substantial, and it need not have made any difference to the outcome, which is the point on which the reform campaign has concentrated.
An intention to assist or encourage. D2 must intentionally assist or encourage, not merely happen to. A person who did not intentionally assist anyone is not liable as an accessory, however close they stood, and nobody can intentionally assist something they knew nothing about. This is the element Jogee restored to English law.
Knowledge of the essential facts. D2 must have known any existing facts necessary to give D1's conduct its criminal character. Someone who helps move a bag without such knowledge of what is in it is in a different position from one who knows.
Where the offence requires a specific intent, an intention that D1 should act with it. Murder requires an intention to kill or to cause really serious harm, which in practice means grievous bodily harm: serious injury, not a bruise. Intending some lesser injury is not intending serious injury, and a jury must be sure the secondary party intended serious injury or death. A secondary party is therefore guilty of murder only where they intended that the principal should act with that particular intent. If they did not, but intended some unlawful violence, the verdict available is manslaughter.
That last point matters more than any other in practice, because it is the difference between a life sentence and a determinate one, and it is decided on evidence about what a defendant knew and meant, almost all of which is gathered in the first weeks of an investigation.
Under Chan Wing-Siu, where two people set out on a criminal venture and D2 foresaw that D1 might commit a second, more serious crime in the course of it, D2 was guilty of that second offence if D1 committed it. Foresight of a real possibility was treated as equivalent to intention. Parasitic accessory liability governed English law for more than thirty years, and it meant a trial judge directed juries on a lower threshold than intention in every joint enterprise case of this kind.
In R v Jogee [2016] UKSC 8, heard by the Supreme Court together with the Privy Council appeal Ruddock v The Queen [2016] UKPC 7 and decided on 18 February 2016, the Supreme Court held that the previous case law had been wrongly interpreted and that the law had taken a wrong turn. The Court said the Chan Wing-Siu principle rested on an incomplete and in some respects erroneous reading of the previous case law, coupled with generalised and questionable policy arguments. Those questionable policy arguments had been repeated for decades without real scrutiny. The correct approach, and the one that applied before 1984, is that foresight is evidence from which a jury may infer an intention to assist or encourage, and is not a substitute for it. A secondary party does not necessarily intend a second crime merely by foreseeing that it might happen, and a person who foresees something does not necessarily intend it.
The practical consequence is a change in what a jury is asked. The question is no longer whether the defendant realised that something might happen. It is whether the defendant intended to assist or encourage it. Such foresight is a fact the jury may weigh in answering that question, and sometimes it is powerful evidence, particularly where a defendant knew a weapon was being carried. It is not the answer.
The principles set out in Jogee had a positive effect on how juries are directed. The Court made clear, though, that putting the law right did not render invalid the convictions reached under the old approach, which is why so many of them still stand. Jogee himself was retried, acquitted of murder, convicted of manslaughter in September 2016 and released in June 2017.
Four things, and each of them is the reason a joint enterprise case can still end in a murder conviction for someone who touched nobody.
Conditional intent still convicts. If the agreement was that serious violence would be used if it became necessary, and it became necessary, that is an intention. It is not foresight of a risk; it is a plan with a condition in it, and it can be tacitly agreed rather than spoken. A common intention of that kind, proved from conduct, is why most murder cases of this type still end in conviction. Proving that common intention is where the prosecution focuses its effort. Where such an intention is proved, the old rule is not needed, and the common intention does the work.
Weapon knowledge remains powerful evidence. The Supreme Court was explicit that knowing an accomplice was carrying a knife is not conclusive of anything, but it will often be very relevant evidence, and in front of a jury it is heavy. It usually becomes the centre of the case.
The contribution still does not have to be significant. Nothing in Jogee introduced a threshold of materiality. Assistance or encouragement of any degree is enough, provided the intention is there.
Old convictions stand unless substantial injustice is shown. This is the most consequential of the four. In R v Johnson [2016] EWCA Crim 1613, the Court of Appeal considered a group of appeals brought out of time by defendants convicted under the pre-Jogee direction. A person seeking exceptional leave to appeal out of time must show "substantial injustice", and the court held that the mere fact the law has since been corrected does not establish it. Where a trial judge faithfully applied the law as it stood, and the jury faithfully applied the direction given, and the trial was fair, the conviction will not be set aside simply because the direction given to the jury is one no judge would give today. Seeking exceptional leave on Jogee alone has therefore rarely succeeded.
The same threshold governs the alternative route. The Criminal Cases Review Commission can refer a conviction to the Court of Appeal where there is a real possibility it will not be upheld, and it has said that only in the rarest of circumstances will a Jogee-based application clear the bar. Murder convictions returned before 2016, where the law was faithfully applied at the time, have very largely stood.
So the correction was, for all practical purposes, prospective. Jogee changed what happens in trials from 2016 onwards. It did very little for those convicted before it.
The joint enterprise cases that cause the most public disquiet are the ones built on association. The prosecution case is that a group set out together, that the defendant was part of it, and that the defendant must therefore have shared a common purpose. The evidence offered is rarely a statement of intention. It is more often a combination of the following.
Phone and cell-site material. Who contacted whom, when, and where each handset was. This establishes presence and association, and it is often presented in a way that implies agreement.
Social media and music. Drill lyrics, videos and group images have been admitted in a number of these prosecutions as evidence of a shared mindset or of gang affiliation. Their admissibility is fought case by case, and the argument that they are prejudicial rather than probative is a real one.
Previous association. Earlier incidents involving the same people, advanced as background evidence of a common purpose.
Movement before and after. Travelling to a location together, leaving together, disposing of clothing, is used to found an inference of a shared plan.
None of this is improper evidence in itself. The difficulty in such cases is cumulative: evidence of both association and presence can be presented as though it proves what each defendant intended, when intention is a separate question that each item only indirectly touches. In such circumstances the answer is not rhetorical. It is the detailed, unglamorous work of showing what each piece of relevant evidence actually establishes, and where the inference breaks.
A person who has assisted or encouraged an offence can escape criminal liability by withdrawing before it is committed, and the law on this is less generous than people expect. Communicating a withdrawal is usually necessary, and in cases of spontaneous violence it may need to be unequivocal. Where the assistance has already been given, for example a weapon has been handed over, something more than walking away may be required.
The wider principle is that a defendant is not liable where their conduct has become so distanced in time, place or circumstances from the principal's conduct that it would be unrealistic to regard the offence as assisted or encouraged by it. Whether that wider principle is satisfied in a particular case is a question of fact, and the evidence that answers it, such as messages sent, a journey taken, a phone switched off, is recoverable early and often lost by the time anyone thinks to look for it.
A prosecutor deciding how to put a multi-handed case has more than one route, and the choice between the possible offences changes the shape of the defence entirely.
Conspiracy, under section 1 of the Criminal Law Act 1977, requires an agreement that a course of conduct be pursued which will necessarily amount to an offence. It is complete on the agreement, whether or not anything happens.
Encouraging or assisting crime, under sections 44 to 46 of the Serious Crime Act 2007, covers conduct intended to encourage or assist an offence, and conduct done believing an offence will be committed. The section 46 offence, dealing with encouraging or assisting one or more of a number of offences, is notoriously broad.
Secondary liability under the 1861 Act requires that the offence was in fact committed.
These overlap, and the choice between them is made at the charging stage. That is a decision a suspect's representatives are entitled to address before it is taken.
A secondary party is "liable to be tried, indicted, and punished as a principal offender". For murder, that means the mandatory life sentence, with a minimum term set on the statutory starting points. The role played is reflected in the minimum term rather than in the sentence itself, and the reduction available for a lesser role is smaller than most people assume.
This is the proportionality problem at the centre of the public argument. Someone who drove a car, or was present and did nothing, and whom the jury found intended to assist serious violence, is sentenced exactly as the primary offender is, even though the primary offender committed murder with their own hands. There is no statutory route to a lesser offence once that finding is made. Criminal liability does not come in degrees at the point of verdict: a defendant is either criminally liable for the offence committed, or not guilty of it.
The Joint Enterprise (Significant Contribution) Bill, introduced by Kim Johnson MP and drafted with Felicity Gerry KC among others, would have amended section 8 of the Accessories and Abettors Act 1861 so that only someone who committed an offence, or made a significant contribution to its commission, could be criminally liable. It received a second reading on 2 February 2024, attracted support across the House but was opposed by the Government, and fell when the 2023-24 session ended.
The argument has not gone away. Kim Johnson tabled the same test as a new clause to the Crime and Policing Bill at Commons report stage in June 2025. Peers returned to it on 5 February 2026, when Lord Garnier moved a probing amendment on joint enterprise in Committee. Neither carried, and the Crime and Policing Act 2026 received Royal Assent on 29 April 2026 without a significant-contribution test in it.
The Law Commission is not the route either, at least not directly. Its consultation on homicide, published in June 2026, expressly declines to propose reform of the general rules of complicity, on the ground that those rules run across the whole of the criminal law rather than homicide alone. What it suggests instead is that a restructured framework of homicide offences might in some cases allow a clearer distinction between the culpability of a principal, who committed the fatal act, and that of an accessory. That consultation closed on 30 September 2026, a further consultation on defences and sentencing is to follow, and no date has been given for a final report.
So no significant-contribution test has been enacted, the Supreme Court's decision stands, and English law on criminal liability for accessories remains as Jogee left it. Anyone reading that a significant contribution is now required is reading about a proposal, not the law.
The evidence base is the reason the argument persists, and it belongs in any honest account of how joint enterprise operates. The CPS's national joint enterprise monitoring scheme recorded 168 homicide and attempted homicide cases prosecuted on this basis in 2024/25, involving 511 defendants. Six in ten were under 25, and one in five was under 18. The CPS's own 2023 pilot found that Black people made up 30% of defendants in its joint enterprise caseload, against 4% of the population. That is a matter of record rather than advocacy.
Not in the closing speech. By the time a multi-handed case reaches a jury, the evidence has been fixed for a year or more, the defendant has usually given an account in interview that has been read back to them many times, and the room contains several other defendants whose cases may cut against theirs. The decisive work happens earlier, and it has a particular shape.
In joint enterprise cases the first interview is frequently conducted while the suspect believes they are a witness to someone else's crime. They explain where they were, who they were with and what they saw, honestly, and in doing so establish presence, association and knowledge: three of the four things the prosecution needs. The decision on whether to answer questions, and what the account should cover, is the single highest-value decision in the case, and it is taken within hours. Our guide to no comment versus a prepared statement sets out how that choice is actually made, and the practical steps to take before an interview under caution covers the preparation.
What the police actually have about this suspect, as distinct from what they have about the group. In multi-handed investigations the two are regularly conflated, and separating them is the first substantive task.
For each item of evidence: what it establishes about this suspect's conduct, and what it establishes about the mental element. The second column is usually much shorter than the first, and demonstrating that is the defence. Material that proves association is not the same as material that proves a common purpose, and the two are regularly presented as though they were. Where that analysis leads to a gap on intention, the gap is the case.
Where the analysis leads to the conclusion that intention cannot be proved against a particular suspect, that is a submission to make to the CPS before a decision is taken, not an argument to run at trial two years later. The Code for Crown Prosecutors expressly contemplates material being submitted by a suspect or their representatives before charge, as set out in our guide to how the CPS decides to charge. In a joint enterprise investigation, the realistic question is often not whether anyone will be charged, but who will be, and that is precisely the kind of decision representations can affect.
I act for people drawn into multi-handed investigations at the stage where the outcome is still open: before the interview, before the charging decision, and before the case has set into a group narrative that is difficult to step out of. The work is pre-charge representation, and it is where I believe these cases are won.
It begins with disclosure, and with separating this client's position from the group's. It continues with the interview decision, taken with that disclosure in view and with a clear understanding of how an innocent explanation of presence can hand the prosecution its conduct element. Where the evidence of intention does not reach the standard, I put that to the officer in the case and to the CPS in writing, with the analysis that supports it, before the charging decision is made.
Where a charge has already been brought, the same analysis goes to the prosecution, seeking a review of the charging decision and, where the evidence does not support the charge, a discontinuance, rather than waiting for a trial to reach the same conclusion at far greater cost. These are also among the few murder cases where the distinction between murder and manslaughter is live on the evidence rather than on the plea, and I prepare that distinction from the outset. This work sits within our wider serious crime defence practice.
The young people who face these allegations frequently have no previous convictions, and are frequently the least involved in the room. That is not mitigation to be raised at sentence. It is the case, and it is made early or not at all.
We offer confidential consultations, in person at our London offices or remotely by secure call. Initial enquiries are handled discreetly and quickly.
If you have been arrested or invited to an interview about an incident involving other people, speak to us before you answer any question. An honest account of where you were, given without advice, is the most common way someone becomes a defendant in a joint enterprise case.
Joint enterprise is not an offence or a separate doctrine. It is the ordinary law of secondary liability, under which anyone who aids, abets, counsels or procures an offence is tried and punished as a principal offender, under section 8 of the Accessories and Abettors Act 1861. The phrase is commonly referred to when several people are prosecuted together for one offence, and it is commonly referred to in the press as if it were a separate crime. It is not.
In R v Jogee [2016] UKSC 8, the Supreme Court held that the courts had taken a wrong turn in Chan Wing-Siu. The Supreme Court and the Privy Council sat together. Foresight that an accomplice might commit a more serious crime is evidence from which a jury may infer an intention to assist or encourage it, and is not a substitute for that intention.
Yes. A person who intended to assist or encourage a killing, or the causing of really serious harm, is convicted of murder and receives a life sentence, whether or not they inflicted any injury or fired the fatal shot themselves. Derek Bentley and Armel Gnango were both convicted on that basis without firing the fatal shot.
No. Presence alone is not assistance or encouragement. It can amount to encouragement where the evidence shows it did, for example where a defendant was part of a group whose numbers emboldened the principal, which is why an account of simply having been there is rarely as safe as it sounds.
No. The law requires assistance or encouragement, with the necessary intention, and sets no threshold of significance. Proposals to introduce one have been debated repeatedly, most recently in 2026, but none has been enacted.
Only on a high threshold. In R v Johnson [2016] EWCA Crim 1613 the Court of Appeal held that an appeal brought out of time in reliance on Jogee requires proof of "substantial injustice", and that the change in the law does not on its own establish it. Successful applications have been rare, but convictions based on Chan Wing-Siu can still be challenged post-Jogee where the evidence of intention was weak and the old direction made the difference between guilty and not guilty.
Secondary liability requires that the offence was actually committed. Conspiracy, under section 1 of the Criminal Law Act 1977, is complete on the agreement itself, whether or not anything follows. A multi-handed case can be charged either way, and the choice is made at the charging stage.
Yes, but the withdrawal usually has to be communicated, and where assistance has already been given, more than simply leaving may be required. The question is whether the defendant's conduct became so distanced in time, place or circumstances from the principal's that it would be unrealistic to regard the offence as assisted or encouraged by it.
Not without advice. In joint enterprise cases an honest account of presence and association supplies the prosecution with much of what it needs, and it is given at a stage when the suspect usually has no idea what the allegation will become. The decision on whether and how to answer is the highest-value decision in the case.
Lex Vindico Group is regulated by the Solicitors Regulation Authority. We represent individuals nationally across England and Wales in criminal, regulatory and parallel-proceedings defence at every stage, and most decisively at the pre-charge stage.
This article is written by Akram Mula, LLM, Solicitor Advocate and CPS-approved Prosecutor, founder of Lex Vindico Group. It is general legal information about joint enterprise, secondary liability and the decision in R v Jogee, not legal advice on any specific case. For advice on your specific circumstances, contact our team directly.
Tell us what has happened and we will respond quickly and in confidence.
.svg)
We’re proud to be recognised by the UK’s leading legal and professional bodies. With decades of experience across criminal defence, regulatory law, and litigation, our clients trust our lawyers to deliver results with discretion, precision, and integrity.
Your next step could make all the difference. Speak to our expert legal team now and protect your future.