Causality (also referred to as causation) is the relation between an event (the cause) and a second event (the effect), where the first event is understood to be responsible for the second. (Wikipedia)
Let’s say a dog approaches a woman:
- with fangs bared, growling, saliva dripping and in a threatening manner. The woman runs away.
- and barks in a friendly manner, wanting attention. The woman ignores the dog and walks away.
- and barks in a friendly manner, wanting attention. The woman pats the dog’s head.
In thermodynamics, the arrow of time helps us differentiate between cause and effect, that is, given equilibrium conditions, the sum of effects can never have lower entropy than the sum of causes. In the example above, cause and effect are clearly defined but the arrow of time gives three different scenarios.
In many policy studies though, causation is treated very much as in the first example, in that because A happened, B cannot help but react to the stimulant. However, it’s good to bear in mind that there are two other (perhaps more?) relationships where B has a choice whether to respond, where the decision to respond may be influenced by other factors not specified in the scenario.
Borrowing from the method in Logic, where causes are divided into necessary, sufficient and contributory, the effects in the example above are divided into “involuntary response”, “voluntary non-response” and “voluntary response”. (No “involuntary non-response” for the threatening dog assuming that it is logical and reasonable for the woman to fear for her safety).
But the main point I wanted to make is that the choice under the “voluntary” branch possessed by the woman is often easily overlooked and under-appreciated. Take the choice function for granted, and policies might fail to achieve their purpose.
There is as well, the desire to believe in the ideal one-cause-one-effect, but the reality of policies is that a whole host of causes must line-up to produce a desired effect. The favourable conditions that may have supported the previous causation may have gone or exist but in weaker strengths.
Causal relationship is not only influenced by agent’s choice but is also affected by stability. Would the relationship still hold under a different environment? It is not unlikely that a high uncertainty environment can produce a different reaction function.
For anyone trying to learn from history it is therefore very important to keep in mind that the range of choices in front of individuals with a free will appear much larger than people writing the history books will acknowledge.