Concept
Indivisible Harm
Apportionment of liability and contribution among tortfeasors
Indivisible Harm
Fermi, a passenger in Joule's car, suffers a broken collar bone when the car collides with a truck Edison is driving. Both drivers were negligent, and Fermi's damages come to $27,000. Edison and Joule did not act together, probably did not know each other, and may not have known the other was on the road.1 Yet each is liable for the entire $27,000. But for Edison's negligence there would have been no accident, and the same is true of Joule's, so each driver caused all of Fermi's harm rather than half of it, and the common-law courts refused to divide the damages in half.2
Full liability matters most when one defendant cannot pay. If Edison is insolvent, Joule still owes Fermi the whole $27,000, whereas a rule making each driver liable for half would leave Fermi with $13,500 from Joule and nothing from Edison.3 Joint liability places the risk of nonpayment on the wrongdoers and leaves them to sort out their shares afterward, a task the sections on contribution take up.
Restatement (Second) of Torts § 879 states the rule: if the tortious conduct of each of two or more persons is a legal cause of harm that cannot be apportioned, each is subject to liability for the entire harm, whether their conduct is concurring or consecutive.4 Three conditions must hold. Each defendant's conduct must be tortious, each must be a legal cause of the harm, and the harm itself must resist division. When all three are present, the law does not ask what share each defendant contributed.
Courts call defendants like Edison and Joule “joint tortfeasors,” which is loose language, though universally used. They did nothing “jointly” in the way that two defendants who agree to beat a victim do; they acted separately, without agreement. It is the injury that is joint. The phrase means only that each defendant contributed to a single, indivisible injury and is fully liable for it.5 That separates this route from concert of action and common duty, where an agreement or a shared obligation ties the defendants together.
The phrase “cannot be apportioned” describes the injury, not the defendants' fault. A death cannot be one-third caused by one defendant. A broken back cannot be sliced into a seventy-percent back and a thirty-percent back. A barn that burns to the ground in a single fire cannot be half destroyed by each of two negligent neighbors. A factfinder may nevertheless be able to say that one driver was seventy percent at fault and the other thirty, because fault can carry percentages even when harm cannot be divided, and jurisdictions that adopt several liability use those percentages to cap each defendant's exposure.6 Whether percentages of fault should limit what a defendant owes for an indivisible injury is a policy question that later sections take up.
Divisible harm looks different, as a contrast shows.
The divisible-harm rules later in this chapter govern cases like that one.
Section 879 covers both concurring and consecutive conduct because the timing of the defendants' acts does not change the rule: each tortfeasor whose conduct is a legal cause of an indivisible injury is liable for all of it.7 Timing still affects how a case is analyzed, so the two patterns get separate treatment below. Concurring tortfeasors are independent actors whose negligence converges at about the same time on one injury. Consecutive tortfeasors act in sequence, and a later actor's negligence aggravates the harm an earlier actor caused.
Concurring Tortfeasors
Independent actors whose negligence converges on one injury are the paradigm concurring tortfeasors, and the rule asks nothing of them beyond causation. They need not have agreed to act together, shared a duty, or even known of one another.
Hill v. Edmonds shows the rule at work. A truck owner let an unlit vehicle stand in the middle of a highway, and a car collided with it. The complaint against the truck owner had been dismissed, and the appellate court ordered it reinstated. Even assuming the car's driver was negligent, the accident could not have happened had the owner not left the truck there, and where separate acts of negligence combine to produce directly a single injury, each tortfeasor is responsible for the entire result, even though his act alone might not have caused it.8 The owner could not shelter behind the driver's negligence.
That result follows from the elements of negligence. A plaintiff must show that the defendant's tortious conduct was an actual and proximate cause of the harm, and a defendant who satisfies those elements has caused the harm whether or not a second wrongdoer also appears. One court explained that an independent, concurring tortfeasor is held jointly and severally liable not because the defendant answers for the other actors' conduct, but because the injury cannot be divided into separate portions and the defendant's own tortious conduct was an actual and proximate cause of it.9 The same reasoning applies when the other contributor is not even in the case. In a suit against an asbestos settlement trust, an unidentified non-party cigarette company had also contributed to the plaintiff's injury, and the plaintiff could still sue the trust for the full amount of damages for an indivisible injury that the trust was a substantial factor in causing.10
The hardest concurring cases involve defendants whose conduct would alone have produced the entire harm. Strict but-for causation seems to excuse both, because each can say the loss would have happened anyway. The courts refuse that escape. They ask whether each defendant's conduct was a material or substantial factor in the harm, and they hold each such defendant liable for all of it.
Kingston v. Chicago & Northwestern Railway Co. is the standard illustration. Sparks from a railroad's locomotive set one fire, and a second fire of unknown origin also burned. The jury found that each fire was a proximate cause of the destruction of the plaintiff's property, and the court found the finding as to the railroad's fire well supported by the evidence.11 The court held the railroad liable and refused to excuse it because the second fire's source was unknown, which would have favored a wrongdoer at the expense of an innocent sufferer. If each of two wrongdoers could plead the other's wrong as a defense, both would escape and the owner would bear the loss, because the damage cannot be apportioned between them. A plaintiff who proves the origin and course of the defendant's fire therefore need not identify where the other fire came from.
Anderson v. Minneapolis, St. Paul & Sault Ste. Marie Railway Co. took the same position. The court declined to free a railroad whenever a fire of unknown origin merged with its own and either fire alone would have destroyed the property, and it approved an instruction that the railroad was liable if its fire was a material or substantial factor in the damage. The same instruction marks the edge of the rule. The railroad was not liable if only a fire not set by its engines burned the plaintiff out, or if its own fire was not a substantial factor in the loss. A defendant who was not a legal cause of the harm owes nothing under § 879. Two motorcyclists present the multiple-sufficient-cause problem on a smaller scale. In Corey v. Havener they passed a plaintiff's horse and wagon independently, one on each side, and either machine alone would have frightened the horse into bolting. Both riders were held liable.
Full liability is fair even in these cases, because a second wrongdoer's arrival takes nothing away from the first. If one person sets a fire with gasoline and another separately sets one with a blowtorch, and either fire would have burned the house down, neither defendant's fault or causal contribution shrinks because the other also acted. Two people who kill a third have not each half-killed the victim.12 A rule excusing each defendant on account of the other would hand both a windfall at the plaintiff's expense.
On an exam, strangers whose negligence happens to converge raise one question: is the harm a single injury that neither defendant's conduct can be separated from? Whether the defendants acted together has no bearing on the answer.
Consecutive Tortfeasors
When one actor commits a tort and a later actor's negligence makes the same harm worse, § 879 still applies, because it reaches conduct that is consecutive as well as concurring.14 But the two defendants no longer stand in the same position. The first actor's conduct is a cause of everything that follows, while the later actor's conduct causes only what it added. The original tortfeasor therefore answers for the whole harm, including its foreseeable aggravation, and the later tortfeasor answers for the aggravation alone.
Medical care is where this pattern appears most often. A physician who negligently diagnoses or treats a problem is potentially liable for the foreseeable consequences of that negligence and also for the foreseeable consequences of a later practitioner's negligence, and successive negligent providers are generally treated as co-tortfeasors under statutes calling for contribution.15 Some courts go so far as to hold a physician liable for subsequent aggravation even when it comes through another physician's additional malpractice.
Foreseeability explains the rule. An injured patient has to seek further care, that care comes from fallible people, and so a later provider's mishandling is one of the risks the original negligence creates. The first actor cannot treat the later error as an outside event that cuts off responsibility, because the later error operates on a harm the first actor produced.
The limit on the later actor follows from the same causal question that burdens the first. The question for each defendant is what that defendant's own conduct caused, and the later actor's conduct caused nothing that already existed when it acted. Charging the later actor with the earlier harm would make it pay for something it never produced. The Restatement's illustrations, quoted in Mitchell v. Volkswagenwerk, mark the boundary from both sides. In the first, a negligent driver fractures a pedestrian's skull and leaves the pedestrian unconscious in the highway, and shortly afterward a second negligent driver runs over the pedestrian and breaks a leg. The first driver is liable for both injuries, the second only for the broken leg.16 In the second, a rider slips on a railroad's negligently maintained platform and breaks a leg. A month later, still on crutches, the rider is on a bus that is negligently operated and collides with another vehicle, and the rider is thrown to the floor and breaks the other leg. Each company is liable only for the damages from the one fracture it caused.17 The two pairs differ in how the first wrong relates to the second event. The first driver's negligence left the victim helpless in the road, so the second injury grew out of the situation that negligence created, whereas a platform fall a month earlier had no part in a bus collision.
The same split appears in second-collision cases in products liability. A truck rams an otherwise non-defective car, which rolls over; the driver's arm breaks in the first impact, and the driver's neck breaks when the roof collapses in the rollover. The truck driver may be liable for all of those foreseeable damages, while the automobile manufacturer, if the roof design was defective, answers under Restatement (Third) of Torts: Products Liability § 16 only for the increased harm the defect caused, which leaves the broken arm outside its responsibility.18 Where the proof cannot separate what the defect added from what the crash would have done anyway, the manufacturer is liable for all of the harm attributable to the defect and to other causes, and it shares that liability with the truck driver jointly or severally, depending on the jurisdiction's rule.19 The hinge is the one § 879 names: a later actor's exposure reaches as far as the harm resists division from what that actor added.
Because the two defendants owe different amounts, the form of the verdict becomes a practical problem. Suppose the jury in the orthopedist hypothetical returns a general verdict of $90,000 that does not separate the leg damages from the shortening damages. The judge can enter the full amount against the driver but has no number to enter against the orthopedist. The cure is a special verdict form asking the jury to state separately the damages from the original injury and from the added injury, so that the judge can enter judgment for the total against the first tortfeasor and for the added amount against the later one.20
The edge of consecutive liability appears in Specialty Hospital-Gainesville v. Barth, a 2020 decision. A patient treated at the defendant hospital developed paralysis and also a bed sore, and the bed sore was later treated at another facility, where it got worse; the second facility was not sued. The court held that a later actor's negligence does not diminish the original tortfeasor's liability where each actor caused a different injury, and that the first facility's fault could be apportioned against a later actor only where the later actor caused the same injury, rather than a different injury or an aggravation of an existing one.21 A later actor's carelessness therefore neither erases nor shrinks the first actor's responsibility for what the first actor caused, and sharing fault between two defendants presupposes one injury that both caused. Where the later actor's harm is entirely its own, each defendant answers for its own harm, which is the subject of the divisible-harm rules later in this chapter.