Proximate cause (also called legal cause) is the requirement that a defendant's negligent conduct be closely enough connected to the plaintiff's injury that the law will hold the defendant liable. Unlike actual cause, which asks whether the injury happened 'but for' the defendant's act, proximate cause asks whether it is fair to impose liability given how the harm unfolded. The dominant test is foreseeability: a defendant is liable for the foreseeable consequences of negligence but not for freakish, unforeseeable results.

Here's the thing that trips up almost every 1L: proximate cause is not really about causation at all. Actual cause has already done the factual work by the time you get here. Proximate cause is a policy filter — a way for courts (and juries) to draw a line and say, "Yes, the defendant caused this in fact, but the law is not going to stretch that far." Once you internalize that proximate cause is about the scope of liability, every case and every exam hypo becomes readable.

Where proximate cause sits in the negligence analysis

Negligence has four elements: duty, breach, causation, and damages. Causation itself splits into two sub-questions, and you must always address them separately:

1. Actual cause (cause-in-fact): Did the defendant's conduct actually bring about the injury? The default test is the but-for test — but for the defendant's act, would the plaintiff have been injured? (Where multiple sufficient causes exist, courts use the substantial-factor test instead.)
2. Proximate cause (legal cause): Even though the defendant actually caused the harm, is the connection close enough — foreseeable enough — that liability is fair?

If you skip straight to "was it foreseeable?" without first nailing but-for causation, you have skipped a step, and on an exam you have left points on the table. Actual cause first, always. Proximate cause is the second gate.

The two competing tests for proximate cause

American law never fully settled on one formulation, and casebooks lean into the disagreement. The two you must know come straight out of Palsgraf v. Long Island Railroad Co. (1928), the case that dominates this section of every Torts course:

1. The foreseeability / risk-rule test (Cardozo, majority)

Chief Judge Cardozo framed the question as one of duty and foreseeable risk: a defendant is only negligent toward those persons and against those harms that fall within the foreseeable zone of risk created by the conduct. If the plaintiff or the type of harm was not a foreseeable consequence of the negligence, there is no liability — the defendant owed no duty as to that unforeseeable plaintiff. This is the majority approach today, usually stated as: a defendant is liable only for the foreseeable consequences of the negligent act.

2. The direct-cause test (Andrews, dissent)

Judge Andrews's dissent in Palsgraf argued that duty runs to the world at large, and the real question is proximate cause, decided by a looser, multi-factor inquiry: was there a natural and continuous sequence between cause and effect, was the connection direct, was the result too remote in time and space? Andrews openly conceded there is no precise formula — proximate cause is where the law "arbitrarily declines to trace a series of events beyond a certain point." That candor is exactly why Andrews shows up on exams: it captures the policy honesty behind the doctrine.

Why both matter: the Cardozo view treats the issue as duty/scope; the Andrews view treats it as pure proximate cause. Most courts have adopted the foreseeability approach, but many professors want you to run the hypo under both and reach the answer each would give.

Applying the foreseeability test element by element

Foreseeability is not one question — it operates on several dimensions, and separating them is how strong exam answers earn their points:

• Foreseeable plaintiff: Was this plaintiff within the zone of danger created by the negligence? (This is the Palsgraf question — Mrs. Palsgraf was standing far down the platform.)
• Foreseeable type of harm: Was the general kind of injury a foreseeable result of the breach? The type must be foreseeable even if the precise mechanism is not.
• Foreseeable manner of harm: Courts are split and often generous here — the general manner should be foreseeable, but courts do not require the exact sequence to be predicted.
• Extent of harm — NOT required to be foreseeable: This is the eggshell-skull rule. A defendant takes the plaintiff as he finds him. If a minor foreseeable impact triggers catastrophic injury because the plaintiff was unusually fragile, the defendant is liable for the full extent. The type of harm had to be foreseeable; the severity did not.

Intervening and superseding causes — the chain-breaker doctrine

The most heavily tested wrinkle in proximate cause is what happens when something else occurs between the defendant's negligence and the plaintiff's injury. Two terms you must keep straight:

• An intervening cause is any event that occurs after the defendant's negligence and contributes to the harm. Standing alone, it does not defeat liability.
• A superseding cause is an intervening cause that is unforeseeable and therefore breaks the chain of proximate causation, cutting off the defendant's liability.

The test is foreseeability: a foreseeable intervening cause does not break the chain; an unforeseeable one (a superseding cause) does. Work through these recurring categories:

• Ordinary negligence of others (foreseeable — not superseding): If you leave a hazard and a rescuer or a second driver reacts negligently, that follow-on negligence is generally foreseeable. "Danger invites rescue" — rescuers are foreseeable plaintiffs.
• Medical negligence following an injury (foreseeable): If your negligence sends the plaintiff to a doctor who commits ordinary malpractice, that is a foreseeable consequence — you remain liable for the aggravation.
• Criminal acts and intentional torts of third parties (often superseding): Generally treated as unforeseeable and superseding — unless the risk of that crime is exactly what made the conduct negligent (e.g., leaving keys in a car in a high-theft area, or a landlord's failure to provide security).
• Acts of God / extraordinary natural events (often superseding): An unforeseeable natural catastrophe can break the chain; a foreseeable storm typically does not.

The clean framing for your answer: identify the intervening act, ask whether it was foreseeable, and label it. Foreseeable → not superseding → defendant remains liable. Unforeseeable → superseding → liability cut off.

Worked examples

Example 1 — foreseeable extent, eggshell skull. D negligently rear-ends P at low speed. A normal person would suffer a sore neck; P has a rare bone condition and suffers a shattered vertebra. Actual cause: satisfied (but-for the collision, no injury). Proximate cause: the type of harm — physical injury from a car crash — was foreseeable, so proximate cause is met. The extraordinary extent is irrelevant; eggshell-skull rule makes D liable for the full injury.

Example 2 — superseding criminal act. D negligently leaves a ladder unsecured against a building. A thief uses it to break into a third party's apartment and assaults the occupant. But-for cause exists, but the intentional criminal act is an unforeseeable superseding cause — proximate cause is not met, and D is not liable for the assault. Change the facts so D is a security company hired precisely to prevent break-ins, and the crime becomes the foreseeable risk that made the conduct negligent — no longer superseding.

Example 3 — running it under both tests. Take the classic Palsgraf scenario: railroad employees negligently push a passenger holding a package, which turns out to contain fireworks; the explosion knocks over scales at the far end of the platform, injuring a distant plaintiff. Under Cardozo's foreseeability approach, the plaintiff was outside the zone of foreseeable risk — no duty, no liability. Under Andrews's direct-cause approach, you would weigh remoteness in time and space and the directness of the sequence — and likely conclude the connection was too attenuated. A top exam answer states both and shows they converge.

The exam trap

The single biggest mistake 1Ls make is collapsing actual cause and proximate cause into one paragraph. Professors write causation hypos specifically to test whether you can keep them apart. But-for causation can be crystal clear while proximate cause is genuinely contested — that is the whole point of the fact pattern. Address them in separate, labeled steps every time.

The second trap is treating foreseeability as a yes/no gut reaction. Don't just declare "this was foreseeable." Argue it — identify the specific risk the conduct created, then show whether the plaintiff, the type of harm, and any intervening cause fall inside or outside that risk. Proximate cause is often a jury question precisely because reasonable minds differ; a strong answer argues both sides before concluding.

The third trap is forgetting the eggshell-skull distinction. Students see a bizarre, severe injury and reflexively call it unforeseeable. Remember: extent of harm need not be foreseeable, only type. Catch that and you'll separate yourself from the pack.

Related doctrines and how they interact

• Duty. Under Cardozo's approach, proximate cause and duty blur — the "foreseeable plaintiff" question can be analyzed as either. Know which framing your professor prefers.
• Actual cause. The necessary predecessor. No proximate-cause analysis matters unless but-for (or substantial-factor) causation is first satisfied.
• Breach / the risk that made conduct negligent. The foreseeability test ties directly back to breach: you ask whether the harm was within the risks that made the conduct unreasonable in the first place.
• Rescue doctrine ("danger invites rescue"). Rescuers are foreseeable plaintiffs, so a defendant whose negligence prompts a rescue is generally liable for the rescuer's injuries.
• Strict liability and products liability. Proximate cause still applies; the fault element changes, but the scope-of-liability screen does not disappear.

Bottom line: proximate cause is where the law decides how far responsibility should reach. Learn it as a policy filter, master the foreseeability test and the superseding-cause rules, keep it surgically separate from actual cause, and you will handle every causation issue an issue-spotter can throw at you.