In tort and in criminal law, a test of whether an individual's action caused a particular event. 409 P. 2d 340, 341. The test is applied by asking whether the event would have occured "but for," or in the absence of, the individual's act. Since every event has several causes, the test can describe any and all of them, and therefore is widely regarded as being of little or no benefit in assigning civil or criminal liability in many types of situations. For example, the test fails to distinguish those causes that should give rise to legal liability from those causes that are an accepted part of everyday life. Thus, if A, driving a car, hits B, crossing the street, the accident has two causes: A's driving and B's crossing. Whether the accident was A's or B's fault would be impossible to determine without factual investigations beyond the "but for" test. Inquiries of this sort are conducted under more complex legal doctrines such as proximate cause. 25 Harv. L. Rev. 103, 109 (1911).
The "but for" test also fails to define cause conclusively when independent, concurrent causes are present. Thus, if both A and B toss matches into a hayloft, and either one's acts would have produced the resultant fire, applying the "but for" test will define neither actor as the cause when in fact both were. 4 Harper, James & Gray, The Law of Torts ยง20.2 (3d ed. 1996).