The popular policeman and other cases: Causal Reasoning - Wagenaar & Crombag - 2005 - Article

Causes can never be seen, it is only an inference. A causal explanation is always a matter of interpretation of events. Every event has a cause, there is no such thing as an uncaused event. However, there are events for which we can’t think of a cause. First people look for a probably cause inside the causal field that we think is the appropriate one (such as animal behavior or politics). If we cannot find an acceptable cause within that field, it might be in another field that you would not immediately choose. The real cause then seems to have such a low subjective probability, that we call it a coincidence.

Important points

• Every event has a cause. Coincidences are events for which we cannot think of a probable cause.

• The more an event meets our expectations, the less we worry about its cause.

• Causal explanation critically depends on our presumed knowledge of the physical and social world.

• Causal reasoning often proceeds regressively. We tend to reason backwards, from effects to probably causes. We usually see an event, and then try to find out what caused it.

The causal field of intuitive mechanics

Major-event major-cause heuristic

Albert Michotte investigated the causal field of intuitive mechanics. He showed that big effects must have big causes, and this is the core of the intuitive theory of mechanics. In the minds of witnesses, bigger effects require bigger causes.

Robert Zajonc

He said that people observing an event first have a nearly instant emotional evaluation of it, and only then they reason about it. Affective judgements are fairly independent of (and happen before) the sorts of perceptions and cognitive operations that are commonly assumed to be the basis of these affective judgements. Attribution of blame is one of these affective judgements.

‘Cues to causality’

There are five ‘cues to causality’. These are criteria that, if met in a given situation, will lead people to choose something as a cause of an event. The five cues are the following:

• Precedence (a chosen cause precedes the effect which it causes).

• Covaration (causes and effects are always together).

• Contiguity in time and space.

• Congruity (causes and effects are similar in length and strength).

• There are only few alternative explanations available.

Multiple causation

Events may not be the result of one single cause. It can be the result of a coincidence of several causes, that come to be under certain conditions. This is often the case with accidents. Accidents differ from premeditated crimes, because they are usually characterized by a complex interaction of different causes.

Even though a single individual may not be responsible for one entire event, he can still be held responsible for his own actions. Three criteria help determine whether or not a person is guilty, and of what:

• Fatal blow criterion (or causa proxima). This is related to the time order of events. The last contribution of giving the fatal blow is the decisive cause. This is the one for which the responsible person should be held liable.

• Discarded insight. The individual who had the most complete insight in the causal forces, but still continued to contribute to the event, has to be held responsible.

• Culpability. This holds responsible any person in the event who had a clear intention to do harm, even when he is not the sole cause.