Actual cause, also called cause in fact, is the factual link between a defendant’s conduct and a plaintiff’s injury, and courts most often prove it through the but-for test: would the harm have happened but for what the defendant did? If the answer is no, the defendant’s conduct is an actual cause. If the answer is yes — the injury would have occurred anyway — the negligence claim fails at the causation stage, no matter how careless the defendant was.1Legal Information Institute. But-For Test
What Actual Cause Means in a Negligence Case
Every negligence claim requires two kinds of causation: factual cause (actual cause) and proximate cause.2Legal Information Institute. Cause Actual cause comes first. It poses a straightforward factual question with no policy weighing attached: did the defendant’s conduct physically contribute to what happened?
This stage works as a gatekeeper. If the defendant’s behavior had no physical impact on the outcome, the case ends before any deeper analysis. A driver who ran a red light three miles from an unrelated crash may have been negligent, but that negligence did not cause the plaintiff’s injuries. Actual cause filters out exactly those disconnected scenarios.
Proximate cause, the second hurdle, asks a different question entirely: even if the defendant’s conduct was a factual cause, should the law hold them responsible for this particular consequence? That inquiry deals with foreseeability and the scope of liability. Many cases never get that far, because they fail the first test.
How the But-For Test Works
The but-for test asks: “but for the existence of X, would Y have occurred?”1Legal Information Institute. But-For Test If the plaintiff’s injury would not have happened without the defendant’s conduct, that conduct is an actual cause. If the injury would have occurred regardless, the defendant is off the hook on causation.
The logic is intuitive in single-actor accidents. A driver runs a red light and hits a pedestrian in the crosswalk. Had the driver stopped, the collision would not have occurred. The failure to stop is the but-for cause. A surgeon operates on the wrong knee. Had the surgeon followed the chart, the healthy knee would be intact. The mistake is the but-for cause.
By mentally removing the defendant’s conduct and asking whether the same result follows, the court isolates what actually mattered in the chain of events. The exercise strips out coincidences and background conditions that had nothing to do with the harm. Juries can apply it without specialized knowledge, which is part of why it has remained the default framework.
When the But-For Test Breaks Down
The test works cleanly when one act leads to one injury. It falls apart when two independent forces each would have been enough to cause the entire harm on their own. In those situations, the but-for test produces an absurd result: neither defendant is the cause, because the harm would have happened anyway due to the other force. Courts call these “merged causes.”3Legal Information Institute. Merged Causes
The classic illustration involves two separate fires, each set by a different person, that converge and destroy a house. Neither fire alone was necessary for the destruction, because the other would have done the job independently. Under a strict but-for reading, both defendants walk free.
The Substantial Factor Test
To avoid that outcome, courts apply the substantial factor test. Each defendant’s conduct is an actual cause if it was a substantial factor in bringing about the harm.4Legal Information Institute. Substantial Factor Test The conduct does not need to be the sole or primary cause. It just has to be more than trivial.
In the two-fires example, both arsonists are liable under this test because each fire was a significant contributor to the result. When both parties breached a duty and both breaches could have independently produced the harm, courts can hold them jointly and severally liable, meaning the plaintiff may collect the full judgment from either defendant.3Legal Information Institute. Merged Causes
When You Can’t Tell Which Defendant Caused the Harm
Some cases present a different problem: multiple defendants acted negligently, but only one actually caused the harm, and the plaintiff cannot prove which one. A strict but-for framework would let every defendant escape, because the plaintiff cannot point to a specific one and say “you did it.” Courts have developed doctrines to prevent that result.
The best-known comes from a case where two hunters negligently fired their guns in the plaintiff’s direction and one bullet struck him. Because it was impossible to determine whose bullet caused the injury, the court shifted the burden of proof to both defendants, requiring each to prove they were not responsible.5Legal Information Institute. Doctrine of Alternative Liability If neither could, both were held liable. The doctrine applies when every potential defendant acted negligently and the plaintiff’s inability to identify the specific cause is not the plaintiff’s fault.
Courts have extended similar burden-shifting logic to mass-produced products where identifying the specific manufacturer is impossible. In one line of pharmaceutical cases, individual plaintiffs who could not identify which company made the pills they took were allowed to recover based on each manufacturer’s share of the relevant market. A company with 30% of the market in the plaintiff’s area bore 30% of the damages.
Actual Cause Versus Proximate Cause
Passing the actual cause test does not end the causation inquiry. A defendant’s conduct can be a factual cause of harm and still not create liability if the connection is too remote. Proximate cause asks whether the harm was a foreseeable consequence of the defendant’s conduct.6Legal Information Institute. Proximate Cause It draws a line around how far liability should extend.
The point is easier to see through an extreme example. Every person who has ever been born is, in the strictest factual sense, a but-for cause of any tort they later commit. A defendant’s mother is a but-for cause of any harm the defendant inflicts. No court would hold her liable, because her act of giving birth was not the proximate cause.2Legal Information Institute. Cause The connection is too attenuated.
Foreseeability is the usual test. If a reasonable person in the defendant’s position could have anticipated the general type of harm that occurred, proximate cause is satisfied. The defendant does not need to foresee the exact injury or the precise chain of events, only the general category of risk.
Intervening and Superseding Causes
An intervening cause is an event that happens after the defendant’s negligence but before the plaintiff’s injury, potentially breaking the causal chain. Whether it actually breaks the chain depends on whether it was foreseeable.7Legal Information Institute. Intervening Cause
A foreseeable intervening event leaves the original defendant on the hook. A landlord who fails to install working locks cannot escape liability when a burglar enters and injures a tenant. Burglaries in unsecured buildings are exactly the risk the landlord’s negligence created. The burglar’s criminal act is an intervening cause, but a foreseeable one.
A superseding cause is different. It is an intervening event so unforeseeable that it replaces the defendant’s negligence as the legal cause. Picture a mechanic who negligently fails to repair a car’s brakes, and while the driver is on the road, a small aircraft crash-lands onto the car. The aircraft strike has nothing to do with the brakes. It is so far outside what anyone would anticipate that it supersedes the mechanic’s negligence entirely.
Defendants routinely argue that some third-party action or natural event broke the chain. Courts evaluate those arguments by asking whether the defendant should have anticipated the type of intervening event, not the exact event. A hospital that discharges a confused patient onto a busy highway is liable when the patient is struck by a car, because that is precisely the risk that made the discharge negligent. The specific driver does not matter.
Proving Actual Cause in Court
If you are the plaintiff, you carry the burden of proving actual cause by a preponderance of the evidence. The jury must conclude that the defendant’s conduct more likely than not caused the harm.8Legal Information Institute. Preponderance of the Evidence Anything above a 50% probability satisfies the standard, which is well below the “beyond a reasonable doubt” standard used in criminal cases.
Lawyers build the factual timeline through physical evidence: surveillance footage, skid marks, medical records, accident reconstruction data. In straightforward cases, the evidence speaks for itself. A rear-end collision at a stoplight with dashcam footage usually needs little more than the video and the medical bills.
Complex cases are different. When the causal mechanism is not obvious, expert witnesses become essential. A biomechanical engineer might explain how a particular impact produced a spinal injury. A toxicologist might link a chemical exposure to a specific disease. Those experts charge significant hourly fees, and hiring them is often one of the largest costs in a tort case.
The Fight Over Expert Testimony
Not all expert opinions reach the jury. Federal courts and most state courts require the trial judge to evaluate whether an expert’s methodology is reliable before allowing the testimony. Under Federal Rule of Evidence 702, an expert may testify only if the proponent shows that the testimony rests on sufficient facts, uses reliable methods, and reliably applies those methods to the case.9Legal Information Institute. Federal Rules of Evidence Rule 702 – Testimony by Expert Witnesses
Reliability is judged on factors like whether the theory has been tested, subjected to peer review, has a known error rate, and has gained acceptance in the relevant scientific community.10Legal Information Institute. Daubert Standard A doctor who testifies that “any exposure” to a toxin causes disease without supporting epidemiological data may be excluded entirely. When the plaintiff’s causation case depends on expert testimony and the court excludes it, the defendant can seek summary judgment, which the court grants when no genuine factual dispute remains.11Legal Information Institute. Federal Rules of Civil Procedure Rule 56 – Summary Judgment
Many causation cases are decided here, long before trial. The but-for test is simple to state. The battle over whether the plaintiff’s expert gets to explain it to a jury is where the case is usually won or lost.